regulation fd a review and synthesis of the academic literature

Regulation FD: A Review and Synthesis of the Academic Literature

• Accounting Horizons 27(3)

• University of Virginia

• University of Texas at Austin

Discover the world's research

• 25+ million members
• 160+ million publication pages
• 2.3+ billion citations

No full-text available

To read the full-text of this research, you can request a copy directly from the authors.

• Account Res J

• Brian J. Bushee
• Michael J. Jung

• J ACCOUNT RES

• Peter D. Wysocki
• J CORP FINANC

• Cheng Xiang
• Qianqian Du

• S.P. Kothari

• K. Sivaramakrishnan

• REV ACCOUNT STUD
• Marcus Kirk
• Zhenhao Jeffery Piao
• Manag Finance

• Gil Aharoni
• Lubomir P. Litov
• Jordan Neyland
• Badryah Alhusaini

• Hal D. White

• J ACCOUNT ECON

• Matthew Gustafson
• Kevin Pisciotta
• Artur Hugon
• An-Ping Lin
• Stanimir Markov

• Tyler J. Kleppe
• Andrew T. Pierce
• Account Horiz

• H. Scott Asay
• Shana Clor-Proell
• Michael Thomas Durney

• A. J. Ostaszewski
• M. H. G. Schröder

• Zhejia Ling
• Jonathan A. Milian
• Antoinette L. Smith
• Elio Alfonso

• Dan S. Dhaliwal

• J Account Publ Pol

• ACCOUNT REV

• Benjamin Segal

• Rafal Szwejkowski

• Dawn A. Matsumoto

• Young Jun Cho

• J FINANC QUANT ANAL

• Kumar Venkataraman

• Jennifer Francis

• Andreas Gintschel

• J. Harold Mulherin
• Charles D. Collver

• Sahiba Chadha

• Chiraphol N. Chiyachantana

• Richard A. Schneible Jr

• Peter D. Easton

• Recruit researchers
• Join for free
• Login Email Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google Welcome back! Please log in. Email · Hint Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google No account? Sign up

Regulation FD: A Review and Synthesis of the Academic Literature

Accounting Horizons, Forthcoming

49 Pages Posted: 19 Oct 2012 Last revised: 13 Jun 2013

Adam S. Koch

University of Virginia - McIntire School of Commerce

Craig E. Lefanowicz

John r. robinson.

Texas A&M University - Department of Accounting

Date Written: March 13, 2013

We summarize the empirical evidence regarding Regulation Fair Disclosure (FD) to gauge whether the regulation achieves its stated objectives and to provide insights and direction for future research. Overall, we find that FD’s prohibition against the selective disclosure of material information eliminates the information advantage enjoyed by certain investors and analysts and thereby provides a more level playing field for all investors. In addition, a number of firms respond to FD by expanding public disclosures and the information environment of the average firm does not appear to be adversely affected. However, we find that an unintended consequence of FD is a reduction in the total amount of information available in the market (i.e., a “chilling effect”) for small or high-technology firms. Finally, ongoing research suggests that private access to management continues to provide select analysts or investors with non-material information used to complete the “mosaic” of information.

Keywords: Regulation Fair Disclosure (Reg FD), Financial Analysts, Earnings Guidance, Capital Markets

JEL Classification: G14, G18, M41

Suggested Citation: Suggested Citation

Adam S. Koch (Contact Author)

University of virginia - mcintire school of commerce ( email ).

P.O. Box 400173 Charlottesville, VA 22904-4173 United States

P.O. Box 400173 Charlottesville, VA 22904-4173 United States 434-924-6356 (Phone)

Texas A&M University - Department of Accounting ( email )

430 Wehner College Station, TX 77843-4353 United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics, related ejournals, financial accounting ejournal.

Subscribe to this fee journal for more curated articles on this topic

S&P Global Market Intelligence Research Paper Series

Subscribe to this free journal for more curated articles on this topic

Accounting, Corporate Governance, Law & Institutions eJournal

Io: regulation, antitrust & privatization ejournal, recommended papers.

Regulation Fd and the Financial Information Environment

By Frank Heflin , K.r. Subramanyam , ...

Regulation Fair Disclosure and Earnings Information: Market, Analyst, and Corporate Responses

By Warren Bailey , Haitao Li , ...

Informational Effects of Regulation Fd: Evidence from Rating Agencies

By Philippe Jorion , Alfred Z. Liu , ...

The Effectiveness of Regulation Fd

By Andreas Gintschel and Stanimir Markov

Who is Afraid of Reg Fd? The Behavior and Performance of Sell-Side Analysts Following the Sec's Fair Disclosure Rules

By Anup Agrawal , Mark A. Chen , ...

The Impact of Regulation Fair Disclosure: Trading Costs and Information Asymmetry

By Kumar Venkataraman , Rex Thompson , ...

Beaver (1968) Revisited: Has the Information Content of Annual Earnings Announcements Declined in the Past Three Decades?

By Wayne R. Landsman and Edward L. Maydew

Managerial and Investor Responses to Disclosure Regulation: The Case of Reg Fd and Conference Calls

By Brian J. Bushee , Dawn A. Matsumoto , ...

Getting Up to Speed on the Financial Crisis: a One-Weekend-Reader's Guide

By Gary B. Gorton and Andrew Metrick

¿En qué podemos ayudarle?

• Co n gresos

Regulation FD : A Review and Synthesis of the Academic Literature.

• Autores: Adam S. Koch , Craig E. Lefanowicz , John R Robinson
• Localización: Accounting Horizons , ISSN-e 1558-7975, Vol. 27, Nº. 3, 2013 , págs. 619-646
• Idioma: inglés
• Texto completo no disponible (Saber más ...)

We summarize the empirical evidence regarding Regulation Fair Disclosure (FD) to gauge whether the regulation achieves its stated objectives and to provide insights and direction for future research. Overall, we find that FD's prohibition against the selective disclosure of material information eliminates the information advantage enjoyed by certain investors and analysts and thereby provides a more level playing field for all investors. In addition, a number of firms respond to FD by expanding public disclosures, and the information environment of the average firm does not appear to be adversely affected. However, we find that an unintended consequence of FD is a reduction in the total amount of information available in the market (i.e., a ''chilling effect'') for small or high-technology firms. Finally, ongoing research suggests that private access to management continues to provide select analysts or investors with non-material information used to complete the ''mosaic'' of information.

Acceso de usuarios registrados

• Dialnet Plus

Opciones de compartir

Opciones de entorno.

• Sugerencia / Errata

© 2001-2024 Fundación Dialnet · Todos los derechos reservados

• Accesibilidad
• Aviso Legal
• R e gistrarse

Regulation FD: A Review and Synthesis of the Academic Literature.

Koch ， S Adam ， Lefanowicz ， E Craig ， Robinson ， R John

展开 

We summarize the empirical evidence regarding Regulation Fair Disclosure (FD) to gauge whether the regulation achieves its stated objectives and to provide insights and direction for future research. Overall, we find that FD's prohibition against the selective disclosure of material information eliminates the information advantage enjoyed by certain investors and analysts and thereby provides a more level playing field for all investors. In addition, a number of firms respond to FD by expanding public disclosures, and the information environment of the average firm does not appear to be adversely affected. However, we find that an unintended consequence of FD is a reduction in the total amount of information available in the market (i.e., a ''chilling effect'') for small or high-technology firms. Finally, ongoing research suggests that private access to management continues to provide select analysts or investors with non-material information used to complete the ''mosaic'' of information. [ABSTRACT FROM AUTHOR]

mosaic theory Regulation FD selective disclosure

10.2308/acch-50500

通过 文献互助 平台发起求助，成功后即可免费获取论文全文。

我们已与文献出版商建立了直接购买合作。

你可以通过身份认证进行实名认证，认证成功后本次下载的费用将由您所在的图书馆支付

您可以直接购买此文献，1~5分钟即可下载全文，部分资源由于网络原因可能需要更长时间，请您耐心等待哦~

百度学术集成海量学术资源，融合人工智能、深度学习、大数据分析等技术，为科研工作者提供全面快捷的学术服务。在这里我们保持学习的态度，不忘初心，砥砺前行。 了解更多>>

©2024 Baidu 百度学术声明 使用百度前必读

The CLS Blue Sky Blog

Columbia Law School's Blog on Corporations and the Capital Markets

Did regulation fd prevent selective disclosure.

The Securities and Exchange Commission proposed Regulation Fair Disclosure (Reg FD) on December 20, 1999. The motivation behind the proposal was concern that an informational advantage provided by selective disclosures to certain market participants was resulting in a loss of confidence in the integrity of capital markets. Thus, the SEC’s stated intention with Reg FD was to “level the playing field” for all market participants.

The proposal was met with a strong reaction by market participants, with over 6,000 comment letters issued in response, and the reaction was mixed. On the one hand, individual investors generally supported the proposal, expressing concern that selective disclosure placed them at a significant disadvantage. On the other hand, analysts and institutional investors were concerned that an unintended consequence of the proposal would be firms reducing their overall disclosure levels, ultimately resulting in less efficient markets.

Early research suggested that Reg FD was effective at reducing selective disclosure without reducing firms’ overall levels of disclosure. However, these studies suffered from significant design weaknesses, and none were able to provide direct evidence that selective disclosure was fully mitigated by Reg FD. Furthermore, the SEC only pursued 10 enforcement actions due to Reg FD violations in the seven year period following its passage, leaving open the possibility that the law has not been well enforced and that selective disclosure could therefore still be taking place.

More recent research on Reg FD overcomes design weaknesses from prior studies and more directly tests whether Reg FD was fully effective at mitigating selective disclosure. We currently have a research study (Campbell, Twedt and Whipple 2016) that uses intraday trading volume and stock returns around Reg FD Form 8-K filings and find that, despite Reg FD’s goal of providing information to all investors simultaneously, disclosure provided pursuant to the regulation appears to be selectively disclosed to subsets of investors beforehand. Specifically, we offer the following set of results. First, we find significant increases in abnormal trading volume during the trading hour immediately prior to the first public release of Reg FD disclosures. In fact, we find that 20 percent of the abnormal volume reaction over the two hour window surrounding Reg FD disclosures occurs during the hour before the disclosure. Second, this pre-disclosure increase in trading volume is larger when the information is of greater consequence to the market. Finally, stock returns during the trading hour immediately prior to Reg FD filings predict returns during the trading hour immediately after the filings, but only for the disclosure of consequential, negative information.

We also find that selective disclosure prior to Reg FD filings is larger for firms with more growth opportunities and weaker information environments. Further, we provide evidence that corporate insiders and large traders appear to be significant beneficiaries of these selective disclosure releases. Specifically, these two groups collectively account for about 50 percent of the abnormal trading volume in the hour leading up to Reg FD filings. However, these results also suggest a role for small traders. While small traders could be unsophisticated investors benefiting from selective disclosure, they could also be large traders splitting up their trades to take advantage of their inside information.

Although our results stand in contrast to those of earlier research on the effectiveness of Reg FD, they are more in line with an emerging literature suggesting that the regulation has not been effective at mitigating selective disclosure. For example, recent studies show that, even after Reg FD, managers have ongoing communication with subsets of investors and that this communication is associated with changes in analyst stock recommendations, abnormal stock price movements, higher trading volume and changes in institutional ownership (Bushee, Jung, and Miller 2011; Green, Jame, Markov, and Subasi 2012; Bushee, Gerakos, and Lee 2012). A limitation of these studies is that they cannot distinguish whether these private communications provide material information that would be a violation of Reg FD, or whether these communications simply provide non-material information that completes the mosaic for these analysts and are thus not a violation of Reg FD. In a literature review of the research on Reg FD, Koch, Lefanowicz, and Robinson (2013) call for studies that are capable of distinguishing between these two explanations. We answer this call by identifying a setting where we know with certainty that the firm intended to disclose information specifically to comply with Reg FD, and we find strong evidence consistent with selective disclosure in this setting.

Bushee, B., J. Gerakos and L. Lee. 2012. Corporate Jets and Private Meetings with Investors. Working paper, University of Pennsylvania, University of Chicago and Boston College.

Bushee, B., M. Jung and G. Miller. 2011. Do Investors Benefit from Selective Access to Management? Working paper, University of Pennsylvania Wharton School, New York University and University of Michigan.

Campbell, J., B. Twedt and B. Whipple. 2016. Did Regulation Fair Disclosure Prevent Selective Disclosure? Direct Evidence from Intraday Volume and Returns. Working paper. University of Georgia and Indiana University.

Green, T., R. Jame, S. Markov and M. Subasi. 2012. Access to Management and the Informativeness of Analyst Research. Working paper, Emory University, University of New South Wales, University of Texas at Dallas and University of Missouri.

Koch, A., C. Lefanowicz and J. Robinson. 2013. Regulation FD: A review and synthesis of the academic literature. Accounting Horizons 27: 619-646.

This post comes to us from Associate Professor John L. Campbell and Assistant Professor Benjamin C. Whipple at the University of Georgia’s Terry College of Business and from Assistant Professor Brady J. Twedt at Indiana University’s Kelley School of Business. It is based on their paper, “Did Regulation Fair Disclosure Prevent Selective Disclosure? Direct Evidence from Intraday Volume and Returns,” which is available here.

• DOI: 10.2139/ssrn.560721
• Corpus ID: 22111752

Re-Examining the Effects of Regulation Fair Disclosure Using Foreign Listed Firms to Control for Concurrent Shocks

• J. Francis , D. Nanda , Xin Wang
• Published 1 June 2004
• Economics, Business
• SPGMI: Compustat Fundamentals (Topic)

173 Citations

The effect of regulation fair disclosure on expectations management: international evidence, externalities of disclosure regulation: the case of regulation fd, how does regulation fair disclosure affect pre-fair disclosure selective disclosers, sec regulation fair disclosure, information, and the cost of capital, regulation fd: a review and synthesis of the academic literature.

• Highly Influenced
• 13 Excerpts

Regulation fair disclosure and the cost of equity capital

Does regulation fair disclosure affect analysts’ forecast performance the case of restructuring firms, disclosure and liquidity management: evidence from regulation fair disclosure, did sox section 404 make firms less opaque evidence from cross-listed firms, 56 references, managerial and investor responses to disclosure regulation: the case of reg fd and conference calls.

• Highly Influential

Regulation FD and the Financial Information Environment: Early Evidence

The effectiveness of regulation fd, a reexamination of information flow in financial markets: the impact of regulation fd and decimalization, earnings and price discovery in the post-reg. fd information environment: a preliminary analysis, who is afraid of reg fd the behavior and performance of sell-side analysts following the sec's fair disclosure rules, stock return volatility before and after regulation fd, discussion of adrs, analysts, and accuracy: does cross listing in the u.s. improve a firm's information environment and increase market value, discussion of adrs, analysts, and accuracy: does cross-listing in the united states improve a firm's information environment and increase market value, related papers.

Showing 1 through 3 of 0 Related Papers

Regulation FD: A Review and Synthesis of the Academic Literature

Chat with Paper

Analysts’ Forecasts and Asset Pricing: A Survey

Corporate jets and private meetings with investors, do investors benefit from selective access to management, do financial analysts restrain insiders' informational advantage, evidence-based policymaking: promise, challenges and opportunities for accounting and financial markets research, information and the cost of capital, is information risk a determinant of asset returns, toward an implied cost of capital, the influence of institutional investors in myopic r&d investment behavior, using analysts' forecasts to measure properties of analysts' information environment, related papers (5), access to management and the informativeness of analyst research, regulation fair disclosure and earnings information: market, analyst, and corporate responses, re-examining the effects of regulation fair disclosure using foreign listed firms to control for concurrent shocks, inside the "black box" of sell-side financial analysts, the effectiveness of regulation fd.

To read this content please select one of the options below:

Please note you do not have access to teaching notes, governance: a review and synthesis of the literature.

Tourism Review

ISSN : 1660-5373

Article publication date: 23 November 2010

Despite the proliferation of the governance concept in the broader academic literature, there is little agreement on definitions, scope and what actually constitutes governance. This is arguably due to the fact that empirical research on the topic, with some exceptions, is generally limited to case studies without use of any common conceptual framework. This is certainly the case in other fields of study and is becoming increasingly obvious in tourism research also. Therefore, the purpose of the paper is to explore and synthesize the governance literature with the objective of identifying the key elements and dimensions of governance.

Design/methodology/approach

Drawing on the two “parent” bodies of literature originating in the political sciences and corporate management fields of study, the paper provides a review and synthesis of the governance concept with the objective of identifying the primary elements and factors that have been employed in studies of governance to date.

A review of 53 published governance studies identified 40 separate dimensions of governance. From this review, the six most frequently included governance dimensions were: accountability, transparency, involvement, structure, effectiveness and power.

Originality/value

A synthesis of the governance literature has not been undertaken to date, either in the tourism literature or in other fields of study, and in doing so the authors provide a basis for tourism researchers to draw on a set of comparable conceptual dimensions in future research. Comparable dimensions which can be replicated and tested in empirical research will add additional depth and rigor to studies in this field.

• Political science

Ruhanen, L. , Scott, N. , Ritchie, B. and Tkaczynski, A. (2010), "Governance: a review and synthesis of the literature", Tourism Review , Vol. 65 No. 4, pp. 4-16. https://doi.org/10.1108/16605371011093836

Emerald Group Publishing Limited

Copyright © 2010, Emerald Group Publishing Limited

Related articles

All feedback is valuable.

Please share your general feedback

Report an issue or find answers to frequently asked questions

Contact Customer Support

Advertisement

Prescriptive analytics systems revised: a systematic literature review from an information systems perspective

• Original Article
• Open access
• Published: 30 August 2024

Cite this article

You have full access to this open access article

• Christopher Wissuchek   ORCID: orcid.org/0009-0001-3384-2007 1 &
• Patrick Zschech   ORCID: orcid.org/0000-0002-1105-8086 2  

Prescriptive Analytics Systems (PAS) represent the most mature iteration of business analytics, significantly enhancing organizational decision-making. Recently, research has gained traction, with various technological innovations, including machine learning and artificial intelligence, significantly influencing the design of PAS. Although recent studies highlight these developments, the rising trend focuses on broader implications, such as the synergies and delegation between systems and users in organizational decision-making environments. Against this backdrop, we utilized a systematic literature review of 262 articles to build on this evolving perspective. Guided by general systems theory and socio-technical thinking, the concept of an information systems artifact directed this review. Our first objective was to clarify the essential subsystems, identifying 23 constituent components of PAS. Subsequently, we delved into the meta-level design of PAS, emphasizing the synergy and delegation between the human decision-maker and prescriptive analytics in supporting organizational decisions. From this exploration, four distinct system archetypes emerged: advisory, executive, adaptive, and self-governing PAS. Lastly, we engaged with affordance theory, illuminating the action potential of PAS. Our study advances the perspective on PAS, specifically from a broader socio-technical and information systems viewpoint, highlighting six distinct research directions, acting as a launchpad for future research in the domain.

Similar content being viewed by others

Prescriptive Analytics: A Survey of Approaches and Methods

From Radical Movement to Organizational Mainstream: ABehavioral Economics Perspective on DSS History

Conceptualizing Automated Decision-Making in Organizational Contexts

Explore related subjects.

• Artificial Intelligence

Avoid common mistakes on your manuscript.

1 Introduction

Decision-making is a cognitive process where individuals select from multiple alternatives. Historically, decisions were primarily based on personal experience, direct observation, or shared knowledge (Santos and Rosati 2015 ). However, with the widespread use of modern information technology, the interconnectedness of society, and the exponentially growing amount of generated data, decision-making has become increasingly complex. As a result, humans began employing mathematical models and algorithms for advanced decision-making to delegate complex decision tasks to computers.

Against this backdrop, business analytics (BA) research to improve organizational decision-making has gained significant traction. The origin of BA is deeply rooted in operations research (OR), commonly linked with decision support systems. Subsequently, the universal adoption of integrated information systems (IS) has enabled organizations to accumulate substantial quantities of data, culminating in the emergence of concepts such as business intelligence (BI) and, more contemporarily, big data analytics (Mikalef et al. 2018 ). Despite the field’s evolving landscape, the core objective of BA has remained steadfast: to delegate analytical tasks to IS to fortify and streamline decision-making processes within organizations. BA shapes well-informed decisions by providing decision-makers with accurate, comprehensive, and timely information, ultimately driving organizational performance and fostering a competitive advantage in the ever-changing business environment (Holsapple et al. 2014 ; Mikalef et al. 2020 ).

Prescriptive Analytics Systems (PAS) embody the most advanced iteration of IS utilized within BA and surpass the capabilities of descriptive analytics, which focus on understanding historical data, and predictive analytics, which forecast the likely future. PAS are designed to guide the best action, considering various factors and constraints to achieve desired outcomes. These systems leverage descriptive and predictive analytics results as inputs, harnessing the insights derived from past events and probable future scenarios to inform their recommendations (Lepenioti et al. 2020 ).

In recent years, research activity has gained traction, with various technological innovations significantly influencing the design of PAS, foremost machine learning and artificial intelligence (AI), such as deep learning, reinforcement learning, and biologically inspired algorithms (Lepenioti et al. 2020 ). Survey papers keep track of these developments by classifying and conceptualizing PAS aspects from different perspectives. However, their primary focus is often on algorithmic facets. Further, analytics has been predominantly viewed as a passive tool to be used by the human decision-maker (e.g., Frazzetto et al. 2019 ; Poornima and Pushpalatha 2020 ; Lepenioti et al. 2020 ).

Recently, there has been a noticeable shift in BA research, with researchers and practitioners increasingly examining the broader implications of analytics systems. The synergistic relationship between analytics or AI systems and their users in decision-making processes has emerged as a critical area of investigation (e.g., Rzepka and Berger 2018 ; Niehaus and Wiesche 2021 ; Hinsen et al. 2022 ). The dynamics in the relationship between analytics tools and human decision-makers are changing drastically, with increased delegation between the two and shifting responsibilities, with analytical systems taking agency and ownership of critical steps in the decision-making process. Specifically, PAS demonstrate a significant bidirectional relationship and an expansive decision-making latitude compared to other analytical systems, which are rather passive, reactive, or anticipatory. Such prescriptive agents can act as human partners or substitutes for behavior-based or outcome-based decision-making (Baird and Maruping 2021 ). We argue that comprehending these factors within the IS community is imperative for illustrating the integration of algorithmic or technical elements into an overarching view, especially given the heightened interest in PAS and their anticipated expansion in research and practice, necessitating the consolidation of the existing knowledge.

To this end, the IS artifact is an established theoretical framework underpinned by general systems theory (GST) to describe, design, or examine systems in a broader organizational context (Chatterjee et al. 2021 ). Following socio-technical thinking, an IS artifact comprises two closely interrelated and connected subsystems: the social system, with humans as its central component, and the technical system, encompassing elements such as technical infrastructure, hardware, and software. The subsystems are nested in an open system that receives inputs and produces outputs in an environment (i.e., organizational or industry context). They are in synergy with each other and adaptable, meaning they can change over time. Thereby, IS artifacts are intentionally designed to meet a specific objective (Bostrom and Heinen 1977 ; Niehaus and Wiesche 2021 ; Chatterjee et al. 2021 ).

We endeavor to build on this perspective and advance the understanding of PAS with three research objectives nested along the principles of GST (Chatterjee et al. 2021 ):

First, we aim to consolidate existing literature on PAS, clarifying the essential subsystems, their constituent components, and their interplay and connection to the decision environment. This effort lays the foundation for future research and is crucial in bridging the current knowledge gap. Understanding these aspects is pivotal for successfully deploying PAS, particularly when considering their integration into organizational decision-making processes.

Second, expanding upon the consistent components, we aim to explore the PAS artifacts from a meta-level perspective. We seek to determine whether the literature unveils recurring archetypes or system designs with distinct characteristics, emphasizing the synergy or delegation between the human decision-maker and prescriptive agent and the degree of adaptation (Baird and Maruping 2021 ). Here, we fall back on human decision-making to better understand how PAS can support organizational decision-making.

Third, as a final objective of our study, we seek to discern the action potential, defining a technology’s capabilities to an individual, organization, or industry for a particular purpose. For this, we fall back on affordance theory, which focuses on the action possibilities arising from the relationship between technologies and their users, often referred to as technology affordances (e.g., Anderson and Robey 2017 ; Mettler et al. 2017 ; Effah et al. 2021 ).

In summary, presenting this synthesized perspective, we posit that it paves the way for subsequent investigations into the core dimensions of PAS within a GST framework, which is especially important for the IS community, effectively setting a research agenda.

To achieve this, we conduct a systematic literature review (SLR), adhering to established methods in IS research (Cooper 1988 ; Webster and Watson 2002 ; vom Brocke et al. 2009 , 2015 ). This approach ensures a comprehensive and rigorous examination of pertinent studies, allowing us to derive meaningful insights and identify knowledge gaps in the field of PAS. The paper’s structure follows: Sect.  2 provides the research context, encompassing decision-theoretic foundations, BA, GST, and related studies. Section  3 introduces our SLR methodology, detailing the steps to analyze the relevant body of work. Section  4 presents the results of our review, focusing on the key concepts and findings that emerged from the literature. Section  5 discusses the implications of our results, highlighting potential future research streams and avenues for further exploration. Finally, in Sect.  6 , we offer concluding remarks, summarizing the contributions of our study and its implications for the field of PAS.

2 Background

To facilitate comprehension of the core facets, in the following sections, we examine human decision-making, BA’s role in improving organizational decision-making, and introducing prescriptive analytics in the GST context as an IT artifact. Subsequently, we assess related studies to identify research gaps and underscore the necessity for further investigation.

2.1 Decision-making

Decision-making is fundamentally a biological process rooted in evolution (Santos and Rosati 2015 ), and decision theory as a research area focuses on examining human choice-making. This field is typically divided into two interrelated aspects (Slovic et al. 1977 ): normative theory, which prescribes ideal decision-making behavior, and descriptive theory, which describes actual human behavior. In the context of our research questions, the normative theory is of greater relevance, as it assumes decision-makers adhere to rules for consistent and optimal outcomes under given conditions. In practical terms, specifically in IS research, normative theory aims to develop tools that enhance human decision-making (Straub and Welpe 2014 ).

In this context, human decision-making generally follows a systematic process. While interpretations may vary across domains, the fundamental structure remains consistent (e.g., Simon 1960 ; Svenson 1992 ; Schoenfeld 2010 ; Darioshi and Lahav 2021 ; Darioshi and Lahav 2021 ), also in an organizational setting (e.g., Trunk et al. 2020 ). The process begins with problem identification, followed by alternative generation, evaluation, selection of the most suitable option, decision execution, effectiveness assessment, and iteration for similar problems. This iterative approach constitutes a continuous learning process, yielding increasingly optimized results over time.

Based on the different interpretations, several authors break down the process into overarching phases (e.g., Ren et al. 2006 ; Leyer et al. 2020 ). In our work, we adopt a triphase decision-making process consisting of the stages before, during, and after the decision (refer to Fig.  1 ). Phase 1, evaluation of alternatives, encompasses problem identification, alternative generation, and ranking. Phase 2, decision-making, involves selecting the most appropriate alternative based on the situation, considering the loss and utility of potential consequences, and executing the decision. Phase 3, adaptation and learning, entails assessing the effectiveness of the outcomes in order to modify behavior for subsequent iterations. The triphase process and its components are crucial in designing a PAS and serve as a guiding framework in the synthesis of this study.

General phases of decision-making processes

2.2 Business analytics

The rapid data volume growth in recent times made BA and Big Data Analytics (BDA) central topics in IS and e-business research (e.g., Pappas et al. 2018 ; Mikalef et al. 2020 ; Jensen et al. 2023 ). Analyzing extensive and diverse data can offer organizations a competitive edge, help achieve strategic and tactical goals, and enhance operational performance by optimizing decision-making processes (Holsapple et al. 2014 ; Knabke and Olbrich 2018 ; Oesterreich et al. 2022 ; Shiau et al. 2023 ). However, the sheer volume of big data, coupled with uncertainty and noisiness, renders it non-self-explanatory (Lepenioti et al. 2020 ). Consequently, extracting value from data necessitates sophisticated techniques, processes, and practices. BA, in this context, is a multidimensional and interdisciplinary concept, drawing on technologies from computer science and engineering, quantitative methods from mathematics, statistics and econometrics, and decision-theoretic aspects from psychological and behavior sciences (Mortenson et al. 2015 ).

BA can be conceptualized using domain, technique, and orientation (Holsapple et al. 2014 ). The domain (i) pertains to the context in which BA is applied (e.g., a decision problem in manufacturing). The second dimension, technique (ii), denotes the methods employed to perform an analytics task, such as linear programming or specific AI or ML techniques. Lastly, orientation (iii) characterizes the objective or direction of thought, addressing questions like ‘what does analytics do?’ or ‘why is it performed?’ and can be regarded as the central dimension. A commonly utilized taxonomy to illustrate the orientation of BA applications is the categorization into maturity levels, which consist of three levels based on their potential and complexity: descriptive analytics, predictive analytics, and prescriptive analytics (e.g., Delen and Ram 2018 ; Frazzetto et al. 2019 ; Lepenioti et al. 2020 ).

The three maturity levels create a synergistic relationship, as depicted in Fig.  2 . Descriptive analytics focuses on the past and present by answering questions such as ‘What is happening?’ or ‘What happened?’ utilizing traditional BI techniques like Online Analytical Processing (OLAP) or data mining (Delen and Zolbanin 2018 ). In contrast, predictive analytics anticipates the likely future by addressing the question ‘What will happen?’ and employs ML methods, such as classification and regression models (Lepenioti et al. 2020 ). Prescriptive analytics seeks to identify optimal decisions, recommendations, or actions by tackling the question, “What should be done?” (Delen & Ram 2018 ). This advanced approach employs sophisticated analytics, operations research, and machine learning techniques, including deep learning, mathematical programming, evolutionary computation, and reinforcement learning (Lepenioti et al. 2020 ).

An overview of the synergies and dynamics of descriptive, predictive, and prescriptive analytics (Krumeich et al. 2016 ; Lepenioti et al. 2020 )

The full potential of predictive analytics can only be harnessed when combined with prescriptive analytics, which streamlines decision-making processes proactively. Reducing the time interval between event prediction and proactive decision-making is paramount to maximizing business value. Prescriptive analytics generates well-informed decisions based on the outcomes of predictive analytics, considering the most suitable timing for executing actions preceding the anticipated event. On the other hand, descriptive analytics can be utilized after the event to scrutinize its underlying causes and consequences while operating on diverse timescales for reactive or long-term actions. In this regard, the prompt detection of the current state and precise forecasting of emerging events are crucial factors in mitigating potential losses in business value (Krumeich et al. 2016 ; Lepenioti et al. 2020 ).

2.3 Prescriptive analytics systems as an IS artifact

The concept of an IS artifact remains ambiguously defined (Chatterjee et al. 2021 ). However, consensus suggests that it can direct research, clarify understanding, set boundaries, provide a design framework, and foster novel research perspectives, among other applications (Orlikowski and Iacono 2001 ; Aier and Fischer 2011 ; Chatterjee et al. 2021 ).

One way of conceptualizing an IS artifact is with GST (Kast and Rosenzweig 1972 ; Chatterjee et al. 2021 ). Within this theoretic framework, and drawing upon socio-technical thinking, two principal subsystems emerge: the social and the technical (Sarker et al. 2019 ). The social subsystem is characterized by its components, encompassing individuals with their inherent knowledge, skills, and values, as well as structural facets like organizational hierarchies and reward systems. Conversely, the technical subsystem is described as an assembly of constituent technical components, such as hardware, software, or methodologies, to transmute inputs into outputs, enhancing the performance of an organization (Bostrom and Heinen 1977 ; Chatterjee et al. 2021 ). These subsystems interact in synergy, allowing for the exchange of information to fulfill mutual objectives or purposes. Situated as an open system, an IS artifact is embedded within its environment (i.e., in an organizational or industry context), influenced by external factors, and concurrently impacting its surroundings. Central to its design is the adaptability of its subsystems, ensuring stability amid changes (Kast and Rosenzweig 1972 ; Sarker et al. 2019 ; Chatterjee et al. 2021 ).

Further, Chatterjee et al. ( 2021 ) underscore the significance of examining the interactions between subsystems using an affording-constraining lens. The affordance theory, introduced initially by Gibson ( 1986 ), identifies action possibilities stemming from the relationship between an object and its observer. In IS research, this concept translates to technology affordances, highlighting the potential actions enabled by the relationship between technologies and their users (Anderson and Robey 2017 ; Mettler et al. 2017 ; Leidner et al. 2018 ). Here, affordance can be defined as ”what an individual or organization with a particular purpose can do with a technology” (Majchrzak and Markus 2013 ), further emphasized by Markus and Silver ( 2008 ), who describe them as a user’s interaction potential with a technical object.

The IS artifact as a theoretical concept will guide our SLR, so we elucidate the aspects and their interplay in an exemplary and simplified PAS-supported organizational decision-making problem within manufacturing, precisely, maintenance operations (e.g., Liu et al. 2019 ; Ansari et al. 2019 ; Gordon et al. 2020 ; Wanner et al. 2023 ), as illustrated in Fig.  3 .

IS artifact with exemplary PAS-supported decision-making problem (own depiction based on Bostrom and Heinen 1977 ; Chatterjee et al. 2021 )

At its core, the social subsystem encompasses the individuals involved in the decision-making process, serving as decision-makers responsible for managing and overseeing maintenance processes. Concurrently, the prescriptive agent, as the technical subsystem, comprises the infrastructure, analytics models, and visualization tools to present findings to decision-makers based on inputs from the decision environment. The decision-maker interacts with the technology components. This interplay affords the optimal maintenance schedules to the user, which then can be actioned upon and implemented in the environment, for instance, a production line with multiple machines.

2.4 Related work

Given the significant research interest in the field, several studies have investigated key topics related to prescriptive analytics. Our analysis emphasizes the importance of our research objectives, and we compile findings, including the purpose addressed in the related work (cf. Table 1 ). We considered existing systematic and unstructured literature reviews to ensure a well-rounded understanding.

Previous research has primarily concentrated on the technical subsystem of PAS, analyzing its technology components and affordances or applications. For instance, Lepenioti et al. ( 2020 ) begin their review with an in-depth analysis of predictive and prescriptive analytics methods before outlining challenges and future directions. Meanwhile, Frazzetto et al. ( 2019 ) take a system-oriented approach, emphasizing the various features of PAS, including productivity, infrastructural considerations, and analytical capabilities. Vanani et al. ( 2021 ) focus specifically on employing deep learning algorithms in PAS in the Internet of Things, while Stefani and Zschech ( 2018 ) provide a conceptualization that considers decision theory as a fundamental aspect. They consolidate various perspectives to derive technology components. Lastly, Poornima and Pushpalatha ( 2020 ) adopt an application-oriented approach, providing a comprehensive overview of the usage of PAS in diverse industries. Despite their differences, these studies collectively emphasize the importance of considering various technical factors when developing PAS.

Aside from general reviews of PAS, some studies focus on specific industries or contexts. For instance, Fox et al. ( 2022 ) emphasized the importance of maintenance tasks in wind farms and conducted a PAS review specifically for this industry. Soeffker et al. ( 2022 ) identified unique requirements for dynamic vehicle routing and reviewed relevant literature. Bhatt et al. ( 2023 ) provided a framework for developing PAS in sustainable operations by identifying five application themes. Meanwhile, Kubrak et al. ( 2022 ) explored challenges and suggested areas for future research to enhance the usefulness of prescriptive process monitoring methods. These studies highlight the significance of context-specific considerations and affordances in developing and applying PAS.

In summary, much of the existing research on PAS has been primarily anchored in its technical components and underlying concepts. This review seeks to weave these diverse strands of thought, capitalizing on the foundational works to offer a more holistic perspective. Specifically, we aim to synthesize the current landscape of prescriptive analytics, positioning it as an IS artifact within the broader context of the decision-making process and revealing the delegation of tasks and responsibilities of both the human decision-maker and the prescriptive agent.

3 SLR methodology

In this section, we employ the established SLR methodology in IS research, as outlined by vom Brocke et al. ( 2009 ; 2015 ), incorporating extensions from Cooper ( 1988 ) and Webster and Watson ( 2002 ). This method consists of five phases: (1) definition of review scope, (2) conceptualization, (3) literature search process, (4) literature analysis and synthesis, and (5) research agenda. Further, we take a descriptive approach to show the current understanding of the literature and reveal patterns, trends, or gaps in current PAS research (Paré et al. 2015 ).

In this section, we begin by addressing the review scope and conceptualization of the topic, drawing on the theoretical foundations from the previous section. Subsequently, we introduce the literature search process, followed by an initial analysis of the literature sample. The synthesis results will be presented in Sect.  4 , while Sect.  5 of this paper will cover the research agenda.

To define and present the scope of our SLR, we employed Cooper’s ( 1988 ) taxonomy with six dimensions. Our (1) focus encompasses research outcomes and applications, including mathematical, conceptual, technological, and infrastructural contributions related to using PAS to understand their aspects better. The (2) goal of our review is to integrate GST perspectives by (3) organizing the results conceptually, following the method outlined by Webster and Watson ( 2002 ). We aim for a (4) neutral representation to reveal the current state of PAS-based research. Our review targets a (5) diverse audience, including IS scholars, practitioners, and specialized researchers from the BA community. Lastly, we endeavor to provide a (6) representative coverage of the relevant literature.

3.2 Conceptualization

In our conceptualization, we draw upon the research context and related work and utilize the GST-based IS artifact to guide the organization and structure of our review and its findings. We aim to uncover crucial concepts and elements within PAS to enable a research launchpad. To effectively address our research goals, we will divide our SLR into three distinct foci:

Constituent components: Understanding the constituent components of a PAS is essential. Here, we adopt GST, precisely the notion that IT artifacts can be viewed as transformational models, receiving inputs, transforming them, and generating outputs (Kast and Rosenzweig 1972 ; Chatterjee et al. 2021 ).

System archetypes: We aim to explore the meta-system level of PAS artifacts by building on the constituent elements. We are keen to ascertain if the literature reveals recurrent PAS archetypes or designs marked by unique features, spotlighting the delegation and responsibilities between the human decision-maker and prescriptive agent and the degree of adaptation. To achieve this, we lean into the decision-making process as a lens to better understand how PAS are nested here.

Technology affordances: From an IS perspective, we aim to understand the purposes for which PAS are implemented and applied. Focusing our analysis on industry or industry-agnostic use cases and breaking the investigation into specific technology affordances, we will delve deeper into how PAS contribute to and influence decision-making tasks across diverse industry and organizational settings.

3.3 Literature search process

The third phase of the SLR methodology, the literature search process, consists of three subphases: database search, keyword search, and backward and forward search. We outline our procedure as follows (cf. Figure  4 ).

The literature search process

Initially, we selected interdisciplinary databases such as Web of Science and Scopus, technology-related databases like ACM Digital Library and IEEE Xplore, and AISeL for IS-related outlets. Our search string combined the term ‘prescriptive’ with several core concepts derived from prior survey articles (Stefani and Zschech 2018 ; Frazzetto et al. 2019 ; Lepenioti et al. 2020 ) (cf. Appendix A for details).

This search yielded 2,597 results (date of search: March 30, 2023). Two researchers collaborated to analyze and screen the papers, using existing conceptual and review papers to establish a shared understanding. Inclusion criteria were set (cf. Table 2 ) to consider only papers that explicitly address the overall design of PAS or describe specific PAS elements, components, or properties. For example, this includes a diverse spectrum of studies, such as conceptual (Levasseur 2015 ; e.g., Appelbaum et al. 2017 ), review (e.g., Poornima and Pushpalatha 2020 ; Lepenioti et al. 2020 ), technological/architectural (e.g., Vater et al. 2019 ; Basdere et al. 2019 ), or mathematical papers (e.g., Bertsimas and Kallus 2020 ; Elmachtoub and Grigas 2022 ). By contrast, we aimed to exclude papers that only briefly mentioned prescriptive analytics without providing more detailed descriptions (e.g., Swaminathan 2018 ; Pereira et al. 2021 ). We also excluded non-academic articles. However, we did not limit our search to only high-ranking journals and conferences to ensure a comprehensive SLR. Webster and Watson ( 2002 ) argue that a topic-centric view of the literature is more valuable than a view limited to a few top journals. Further, we excluded articles that were not written in English.

To incrementally exclude irrelevant papers, we applied a stepwise procedure. First, a title and keyword analysis identified 799 relevant articles. After reviewing abstracts and removing duplicates, we removed 449, reducing the number of papers to 350. Full-text screening further reduced the number of relevant articles to 198. To supplement our findings, we conducted citation chaining, both forward search (via Google Scholar) and backward search (via bibliography), adding 64 relevant papers and increasing the total to 262 articles. The complete list of the literature sample is available in Appendix B.

3.4 Overview of the literature sample

Our findings indicate a significant growth in research interest in prescriptive analytics. More than half of our sample was published after 2020, demonstrating the increasing relevance of this area of research. Concerning publication types, approximately 60% of articles are in journals, and 34% of our sample comprises conference papers. A smaller proportion, 6%, is book sections or chapters.

An initial sample analysis allowed us to explore thematic trends by observing the top research outlets. Much of the sample is published in IEEE and ACM proceedings focusing on computer science and technology. Similarly, we observed a clear indication of the operations research and management domain, with publications in journals such as Management Science, European Journal of Operations Research, and others at the intersection between computing, operations research, and industrial engineering (Fig.  5 ).

Overview of literature sample and top research outlets

In 2021 and 2022, we noticed a considerable drop in conference papers, possibly due to widespread lockdowns in light of the COVID-19 pandemic. Nonetheless, the publication of numerous journal papers during this period contributed to the continued growth of interest in those years. In summary, after initial observation of our literature sample, it is evident that the research is heavily weighted toward technological and mathematical disciplines, which is to be expected given the core of prescriptive analytics.

Below, we showcase the findings of the synthesis. We will begin by detailing the constituent components, system archetypes, and finally, the technology affordances of PAS.

4.1 Constituent components

We observed a predominant emphasis on technical components within our literature sample, characterized by its technical focus and the inherent nature of prescriptive analytics. The decision-maker naturally emerges as the pivotal entity in the social subsystem. However, a detailed exploration of the decision processes or structures surrounding PAS-based decision-making is absent in current research, with just a few authors focusing on the decision-maker. For example, Käki et al. ( 2019 ) investigate the deviations of decision-makers from model-based recommendations and their impact on the effectiveness of decision-support processes. By examining these discrepancies and discerning the underlying motivations, the authors emphasize the potential for improving the planning process, optimizing model-driven decision-making, and refining the lifecycle management of PAS. Similarly, Caro & de Tejada Cuenca ( 2023 ) study the adherence to prescriptive analytics recommendations, highlighting trust as a deterrent, with interpretability as a crucial intervention.

Consequently, the following sections will mirror this emphasis on the technical subsystem and its components. Here, we present a multi-layered concept matrix with 23 concepts. The basic structure of the concept matrix follows GST, that IT artifacts can be viewed as transformational models, receiving inputs, transforming and processing them, and generating outputs (Kast and Rosenzweig 1972 ; Chatterjee et al. 2021 ), which we coin “decision formulation”, “decision input”, “decision processing”, and “ decision output” in our study. We added “ancillary components” to address additional aspects (Frazzetto et al. 2019 ) that are not situated in the core of the prescriptive agent but support its integration into organizational decision-making processes and structures.

Table 3 presents a concise summary of the outcomes derived from the concept matrix, featuring exemplary studies corresponding to each concept and the number of hits where the concepts are discussed or mentioned in the text corpus. A detailed overview of all concepts correlated with the literature sample can be found in Appendix C. Further, we added a trend analysis in Appendix H, visualizing the concepts’ development across the years.

Each concept (emphasized in bold) is expounded upon in greater detail in the subsequent sections. Further, Fig.  6 integrates and summarizes the findings, offering a cohesive representation of the PAS while elucidating the interactions among its core components.

Exemplary visualization of constituent components of PAS in a coherent view

4.1.1 Decision formulation

Decision formulation refers to the essential elements for structuring a decision problem, subdivided into decision variables, objectives, and constraints. Decision variables define the object of interest within a decision (Stefani and Zschech 2018 ). For instance, production planning may involve mapping the manufacturing workforce, machinery, and material flow allocation to each other most profitably (Elmachtoub and Grigas 2022 ). In this context, the complete set of all potential mappings constitutes the set of all alternatives or competing decisions. When considering in conjunction with the encompassing environmental conditions and various contextual factors, the specification of decision variables plays a pivotal role in delineating specific states and their corresponding outcomes. These states frequently exhibit associations with utility values, encompassing metrics such as costs, profits, or revenues, thus serving as quantitative indicators for overarching objectives that demand either minimization or maximization. These objectives are commonly denoted as objective functions, optimization functions, or simply objectives. Additionally, it is pertinent to acknowledge the presence of constraints , which often encircle decision spaces, emanating from natural limitations (e.g., capacity limitations of a machine) or managerial imperatives (Stefani and Zschech 2018 ; Elmachtoub and Grigas 2022 ).

4.1.2 Decision input

The input is the foundational component in data-driven decision-making, encompassing essential attributes of analytics processes. While not intrinsic to the core components, numerous authors emphasize these attributes in the context of PAS (cf. Appendix G for an overview). These properties encompass the structural characteristics of the data (Lash and Zhao 2016 ), its origin, whether external or internal (Bertsimas and Kallus 2020 ), and the manner of data generation, whether reliant on human-based assumptions, empirical methods or synthetic means (Stefani and Zschech 2018 ; Ceselli et al. 2019 ). Additionally, consideration is given to the velocity of data, differentiating between historical and real-time data (Krumeich et al. 2016 ; Miikkulainen et al. 2021 ). Subsequently, data undergoes various preprocessing or data engineering steps to prepare the data input for analytical processing (McFowland III et al. 2021 ), often handed over to descriptive and predictive functions, resulting in the current state and probabilities. Many authors see these preceding analytical results as a foundation of PAS (e.g., Wang et al. 2018 ; Miikkulainen et al. 2021 ), thus making them constituents.

Descriptive analytics provide insights into the current state , such as patterns or key performance indicators, to assess the existing conditions within the decision context. This information helps identify areas that require modification compared to the current state and serves as a baseline for evaluating the ramifications, e.g., in terms of gains or losses, of a decision (Li et al. 2021 ). Moreover, it is imperative to recognize that decision problems inherently encompass a degree of uncertainty. This uncertainty may be effectively quantified by utilizing probabilities derived from predictive analytics, facilitating an elucidation of the likelihood associated with the impending occurrence of a particular outcome. Probabilities frequently serve as integral components, directly integrated as weightings within the delineation of the objectives within the overarching decision framework (Stefani and Zschech 2018 ; Wang et al. 2018 ; Lepenioti et al. 2020 ; Miikkulainen et al. 2021 ).

4.1.3 Decision processing

Both the input and formulation frame the decision processing, correspondingly generating prescriptions. Here, the literature refers to various techniques, which can be grouped into mathematical programming, evolutionary computations, machine learning, probabilistic, and logic-based models, which are not mutually exclusive and can be utilized interactively or sequentially, confirming the results of prior literature reviews (Lepenioti et al. 2020 ). Additionally, we refer to Appendix F for an overview of more detailed techniques and potential subcategories.

Mathematical programming is widely adopted for optimizing objective functions within constrained solution spaces (e.g., McFowland III et al. 2021 ). In contrast, evolutionary computations offer bio-inspired optimization techniques (e.g., Miikkulainen et al. 2021 ), while machine learning (ML) algorithms enable learning without explicit instructions (Janiesch et al. 2021 ). Supervised ML facilitates anticipatory decision-making by predicting unseen data (Lash and Zhao 2016 ). As a subset of ML, reinforcement learning (RL) aims to maximize cumulative rewards in given environments, proving effective for well-formalized decision-making problems (Lepenioti et al. 2021 ). Probabilistic models , such as Bayesian inference or Markov models, calculate event likelihoods and represent causal relationships (Lepenioti et al. 2020 ). Logic-based models examine chains of cause-and-effect relationships leading to specific outcomes (Lepenioti et al. 2020 ). Lastly, simulations enable exploring hypothetical or real-life processes to improve decision-making by generating scenarios and uncovering optimal behaviors for specific situations (Lepenioti et al. 2020 ).

4.1.4 Decision output

The decision output refers to the result or outcome produced in decision processing. It represents the prescribed course of action or solution from competing decisions based on the decision variable (e.g., all alternative configurations of human resources, machinery, and equipment). While the optimal or single decision is typically required, authors emphasize the importance of making the alternatives or multiple decisions transparent to the decision-maker and accommodating more complex situations, such as dynamic environments. For example, the prescriptions are tailored to sequential process stages, varying time points, or ever-changing states, ensuring a more adaptive response (e.g., Liu et al. 2019 ; Brandt et al. 2021 ).

4.1.5 Action mechanisms

Following the decision-making process, the human decision-maker traditionally performs the subsequent actions within the decision environment. In this context, it is essential to emphasize that the implications and outcomes are detached from the technical subsystem, which primarily functions as a passive tool at the user’s disposal. However, recent advancements in AI and ML have triggered a notable shift in IS research (Baird and Maruping 2021 ). This shift acknowledges the agency of the IS artifact and promotes synergistic collaboration between the technical and social subsystems. In the context of PAS, this transition necessitates the development of specific mechanisms to facilitate action execution, tracking, and adaptation, enabling the technical components to engage in the decision environment autonomously.

Implementing execution mechanisms is essential to facilitate actions within the decision environment, distinguishing between two primary execution modes: (i) autonomous execution, where the prescriptive agent independently carries out the decision (e.g., Mazon-Olivo et al. 2018 ; Soroush et al. 2020 ), and (ii) execution with human intervention, which may require human confirmation of the decision output before implementation (e.g., Rizzo et al. 2020 ). As automation levels increase, the importance of tracking becomes more pronounced to document the actions taken in the decision environment, leading to adaptation mechanisms to change iteratively by using insights from tracked actions and their outcomes (e.g., Bousdekis et al. 2020 ; Zhang et al. 2021 ),. This adaptation is driven by the implications of decisions made within the decision environment and dynamic shifts in decision inputs, particularly in dynamic and evolving contexts. Also, this can be done autonomously by observing the decision-maker or the outcomes in the decision environment (Liu et al. 2019 ; Tamimi et al. 2019 ), but also with the decision-maker’s input (e.g., Krumeich et al. 2016 ; Kim et al. 2020 ; Vater et al. 2020 ; Miikkulainen et al. 2021 ). Action mechanisms are crucial in distinguishing between the PAS archetypes we identified in our synthesis. Therefore, we will delve deeper into these aspects in the respective Sect. ( 4.2 ).

4.1.6 Ancillary features

In addition to the formulation, input, processing, and action mechanism components, our literature review has revealed further concepts within the context of PAS. We term these concepts "ancillary" since they are not at the core but somewhat secondary to PAS functionality. The literature encompasses many features, from data properties and general infrastructure considerations to integrating PAS within manufacturing systems (e.g., Vater et al. 2019 ; Ansari et al. 2019 ; Consilvio et al. 2019 ). Our focus, however, remains on overarching aspects and concepts that align with our research objectives. Specifically, we concentrate on features directly connecting to how PAS is situated within the organizational context and their role in decision-making processes.

Integration is a central concept determining a PAS’s positioning in the broader (inter-)organizational landscape. Vertical integration allows data and decision outputs to be available across hierarchical levels, while horizontal integration incorporates data throughout processes or business functions, even externally (Appelbaum et al. 2017 ; Kumari and Kulkarni 2022 ). The growing data volume and sophisticated algorithms demand increased computing power, addressed through distributed computing , linking computational resources for shared data and processing power (Lepenioti et al. 2020 ). Modularization reduces complexity by, for instance, separating descriptive, predictive, and prescriptive analytics or specific functions (Appelbaum et al. 2017 ; Frazzetto et al. 2019 ). Additionally, security- and privacy-preserving features , though a niche in current research, are crucial due to rising cyber threats. For example, Harikumar et al. ( 2022 ) propose an algorithm for private prescription vectors.

With a focus on the decision-maker’s perspective, our review has unearthed studies discussing explainability within PAS-based decision-making. The objective is to bolster user trust in the decision-making process, facilitating the adoption and effective implementation of system recommendations in real-world scenarios (e.g., Mehdiyev and Fettke 2020 ; Notz 2020 ; Suvarna et al. 2022 ). Visualization is pivotal as a design feature in PAS, guiding users visually through the decision process. Visualized results prove instrumental in enabling users to swiftly grasp decision outcomes and potential consequences (e.g., Appelbaum et al. 2017 ). The workflow interface serves as a guiding element, allowing users to navigate the decision process with the flexibility to adjust input parameters, underlying models, or output validation. These adjustments can be facilitated through no-code or traditional programming interfaces (e.g., Frazzetto et al. 2019 ). Furthermore, extensibility options are paramount in PAS, allowing users to install or develop components tailored to specific use cases (e.g., Frazzetto et al. 2019 ).

4.2 System archetypes

Per our study’s objectives, this section delves into PAS from a meta-level perspective. The overarching goal is the enhancement of organizational decision-making. Building on the decision phases in the background section, we use these as the foundation to conceptualize archetypes, denoting overarching designs or setups with distinct characteristics. Through this lens, we aim to illuminate the synergy among the technical and the social subsystems, the prescriptive agent and human decision-maker, respectively, underscoring their collective role in refining organizational decision-making processes.

Given the extensive body of literature, configurations of the constituent components are diverse, depending on the industry, application, or specific use case. While many authors primarily emphasize technical aspects, the analysis of this literature reveals recurring patterns in the overall structure of PAS. The patterns are often anchored in the general decision-making phases (evaluation of alternatives, decision-making, and adaptation). One noteworthy observation is that technical subsystems are not uniform in their role within the decision-making process, nor their interaction with human decision-makers, and within the reviewed literature, a discernible shift emerges. Prescriptive analytics is evolving from a passive tool used by human decision-makers to having agency and assuming responsibilities of their own in the decision-making process. Prescriptive agents exhibit a growing decision-making latitude, and they can assume the role of substitutes for behavior-based or outcome-based decision-making by prescribing, autonomously executing actions, and adapting to changes in the decision environment (Baird and Maruping 2021 ).

To conceptualize these findings, we draw upon the theoretical framework of IS delegation proposed by Baird and Maruping ( 2021 ), anchored in agent interaction theories. Specifically, we adopt delegation mechanisms to delineate and explain four distinct system archetypes: advisory , executive , adaptive , and self-governing PAS. Our focus is directed toward understanding the (i) levels of delegation and the (2) roles or responsibilities played by human decision-makers and prescriptive agents across the three decision-making phases. Table 4 summarizes the key characteristics of each archetype, and a complete overview of the identified archetypes in our literature sample is available in Appendix E. Additionally, Table  5 demonstrates the roles and authority of the prescriptive agent and human decision-maker in each archetype. The agency of the other is not always entirely removed, and it rather pertains to the primary responsibility of a delegator or proxy-based relationship (Baird and Maruping 2021 ). We detail this in the following sections, where we will analyze the archetypes identified through our study, supported by exemplary visualization (refer to Figs.  7 , 8 , 9 , and 10 ) of the responsibilities and delegation mechanisms in the PAS-based decision-making process.

Exemplary visualization of delegation mechanisms and responsibilities in advisory PAS

Exemplary visualization of delegation mechanisms and responsibilities in executive PAS

Exemplary visualization of delegation mechanisms and responsibilities in adaptive PAS

Exemplary visualization of delegation mechanisms and responsibilities in self-governing PAS

4.2.1 Advisory PAS

The advisory archetype is notably the most common variant in the literature sample by a significant margin. In this archetype, prescriptive agents contribute only to the initial phase by assessing alternatives and presenting the optimal decision or course of action to the user, who maintains full decision-making authority. These prescriptive agents are static and do not adapt to the consequences of a decision or changing environments, necessitating manual adjustments or reconfigurations of inputs or underlying models by humans. The prescriptive agents are, in this sense, mostly passive tools to be used by the decision-maker with minimal delegation or agency in the decision-making process and entirely disconnected from the problem environment. For instance, Abdollahnejadbarough et al. ( 2020 ) explore a telecommunications provider employing an advisory PAS for supplier management. The system collects data from internal ERP and external supplier sources before employing machine learning to cluster suppliers. Subsequently, an optimization engine processes the results to recommend the most efficient suppliers for sourcing decisions. The decision-maker handles the following steps, such as contacting suppliers or executing purchase orders, without further involvement of the prescriptive agent.

Though less extensively researched, there are some examples in the literature where delegation does happen between the decision-maker and the prescriptive agent during the evaluation of alternatives phase (cf. Figure  7 ). This interaction might involve adjusting inputs or decision variables to accommodate real-world factors, expert knowledge, or risk preferences by the decision-maker. For example, Kawas et al. ( 2013 ) outline a PAS for sales team assignments that recommends optimal allocations while allowing decision-makers to fine-tune output through what-if analyses. This approach incorporates expert judgment (i.e., expert-in-the-loop), such as customer sentiment or subjective preferences, enabling experimentation with diverse sales team configurations.

4.2.2 Executive PAS

Executive PAS is the least common archetype in our sample. With only twelve papers, this archetype represents a niche in current PAS research. Humans traditionally hold the mandate to act upon a prescriptive output. However, the literature also suggests some PAS designs in which the prescriptive agent receives the authority to execute decisions autonomously in the problem environment. In these systems, adaptation remains static, and the prescriptive agent is responsible for the initial two phases, with minimal interaction with the decision-maker. Executive PAS are predominantly utilized in domains with high automation, standardization, or repetitiveness, where rapid decision-making is necessary.

For example, Soroush et al. ( 2020 ) introduce a PAS recommending optimal safety actions to detect and address process operation hazards by implementing mitigative chemical-process measures. Similarly, Mazon-Olivo et al. ( 2018 ) describe a PAS in precision agriculture that autonomously sends repetitive and planned actions to IoT devices in the field. Additional examples include intelligent call center routings where an optimal service employee is matched to a customer (Ali 2011 ) and a data allocation scheme across a Hadoop cluster for enhanced data security and privacy (Revathy and Mukesh 2020 ). In some cases (cf. Figure  8 ), there is a higher degree of delegation, where the prescriptive agent executes the decision in the environment, but a decision-maker must first approve or validate the output (Rizzo et al. 2020 ). This approach can benefit high-stakes decision-making with significant financial implications or safety and compliance concerns.

4.2.3 Adaptive PAS

In the case of an Adaptive PAS, while human decision-makers maintain authority over the decision-making phase, the prescriptive agent assists in the adaptation and learning phases, contributing to a more effective and well-informed decision-making process. The prescriptive agent monitors decision outcomes and their impact on the decision environment, incorporating observations into subsequent iterations by adding new data as input or dynamically adjusting the decision model. As problem environments often change due to shifting requirements, priorities, or new knowledge, adaptive PAS, as a dynamic archetype, holds significant potential compared to static counterparts.

For example, Liu et al. ( 2019 ) propose a system for optimizing locomotive wheel maintenance operations, recommending inspection schedules to minimize long-term cost rates. Similarly, Zhang et al. ( 2021 ) present a reinforcement learning-based maintenance optimization model that determines optimal actions based on a machine’s ever-changing degradation state. Bousdekis et al. ( 2020 ) emphasize the importance of feedback and learning mechanisms in a generic IoT scenario, where an agent suggests optimal actions to users and updates the prescriptive model dynamically based on real-time IoT sensor data. Prescriptive Agents can also observe the decision environment and decision-makers while actioning. Tamimi et al. ( 2019 ) discuss a PAS for field development design, recommending optimal designs and deriving the decision-maker’s utility function for subsequent iterations of prediction and optimization models. Käki et al. ( 2019 ) highlight the deviation from model recommendations in production planning, often resulting in deteriorated performance, and emphasize the added value of the adaptive PAS compared to static counterparts.

There are also examples in the literature where decision-makers observe action consequences or judge potential outcomes based on domain knowledge (i.e., expert-in-the-loop systems). Here, the prescriptive agent requires active input from the human to adapt for subsequent iterations, with decision-makers providing information by, for example, relabeling outputs or aggregating real-world outcomes into the training set for future cycles (e.g., Krumeich et al. 2016 ; Kim et al. 2020 ; Vater et al. 2020 ; Miikkulainen et al. 2021 ). This archetype indicates that expert knowledge and human judgment remain vital in the adaptation and learning phase. Also, from the perspective of GST, adaptation is crucial, as the social and technical subsystems naturally change over time (Chatterjee et al. 2021 ).

4.2.4 Self-governing PAS

The fourth archetype, self-governing PAS, represents a potentially fully autonomous system where the prescriptive agent has agency and responsibilities in the entire decision process independently or with the decision-maker’s involvement. Combining the capabilities of the other three archetypes, the self-governing PAS is the most sophisticated version, offering the highest added business value due to automated execution, adaptability, dynamic self-learning mechanisms, reduced manual work, and enabling rapid, fact-based decision-making in dynamic environments.

Self-governing PAS relates to well-researched and practiced areas such as route optimization and data load distribution (cf., Wang et al. 2008 ; Jozefowiez et al. 2008 ), used in highly structured environments, which could be considered precursors or early manifestations of the archetype. However, they differ from more recent examples in aspects like their integration into the broader infrastructural landscape and their use of historical and real-time data. Moreover, being less disjointed from the decision environment, they support the entire decision-making process while showing a high potential for delegation between the human decision-maker and the prescriptive agent.

Examples from our literature sample primarily come from domains with high technological maturity and sophistication, such as cyber-physical systems, IoT, modern energy distribution systems, or smart-sensor-driven environments. These technologies are inherently data-driven, automated, and integrated—ideal preconditions for advanced agents. For instance, Ceselli et al. ( 2019 ) propose a data-driven framework for optimally distributing data traffic from mobile access points across capacity-constrained mobile edge cloud networks. A fully autonomous orchestrator module executes the best data assignment plans. The selected plans, their effects, and the access points’ demands are continuously logged and validated for subsequent iterations. Similarly, Vater et al. ( 2020 ) introduce an IoT-based architecture for real-time error detection in automotive manufacturing, utilizing edge-/cloud-architecture including modules for preprocessing, prediction, prescription, action-taking, and validation to close the loop for a fully autonomous decision-making process.

Gutierrez-Franco et al. ( 2021 ) present a PAS for last-mile delivery operations as another example. The system leverages historical data such as traffic, customer behavior, and driver performance as input. This data is initially preprocessed and descriptively analyzed, forming a foundation for predicting future operations and prescribing optimal routes or schedules. The generated output is fed into an execution module, providing optimal routes for drivers. Furthermore, real-time circumstances, including traffic or route deviations by drivers, are captured via GPS or the vehicle’s sensors, allowing continuous recalculation of the optimal schedule. In this instance, the decision process is not linear but dynamically adapts based on the current state of the problem environment. At the end of each shift, a learning mechanism initiates, collecting accumulated data and best practices to enhance delivery operations, serving as historical data input for subsequent days.

4.3 Technology affordances

As outlined in the scope of our SLR, our goal is to uncover affordances that represent specific purposes or decision-making tasks supported by PAS. The resulting concept matrix begins with industry (bold text) and is divided into specific affordances (italic text). We use affordances in the sense of action potential. The technical nature of our literature sample poses a challenge in extracting both the perception and actualization of affordance. In this context, we can only derive how the social subsystem perceives the affordance, for example, through visualization. Conversely, actualization can only be detailed based on the actor responsible for executing the decision, either the technical component or the human decision-maker (Pozzi et al. 2014 ; Leidner et al. 2018 ) (cf. Figure  11 for details), which we detail in the previous section on system archetypes.

Affordance theoretical framework mapped to PAS and the decision-making phases (own depiction based on Pozzi et al. 2014 )

Further, our literature sample includes papers that do not address a specific affordance but provide a more general perspective, such as mathematical or algorithmic formulations, infrastructural considerations, reviews, and conceptual papers. Therefore, they are excluded from our considerations in this section. In the following, we will focus on more prevalent industries (N > 5) and affordances, referring to Table  6 for niche examples. We conclude this section with a summary and overarching affordance patterns. Additionally, Appendix D provides a comprehensive overview of all affordances mapped to the literature sample. Further, Appendix H includes a trend analysis across the more prominent industries.

4.3.1 Manufacturing

As the most researched industry, manufacturing has historically relied on mathematical models to optimize processes, workforce allocation, and schedules, which are crucial in enhancing efficiency and productivity. With Industry 4.0, manufacturing has transformed into a highly developed sector integrating advanced technologies like robotics and IIoT to establish intelligent, interconnected production systems (Wanner et al. 2023 ). As reflected in our literature sample, these developments have spurred extensive research interest in data-driven analytics. Maintenance planning , a dominant affordance for PAS, has emerged as a research stream called prescriptive maintenance. Here, the primary purpose is to afford optimal maintenance schedules, often incorporating spare parts management, primarily driven by machine sensor data. For example, these systems employ descriptive analytics to analyze the current machine state and predictive analytics to foresee potential failures, fueling a prescriptive model to propose the optimal schedule (e.g., Liu et al. 2019 ; Ansari et al. 2019 ; Fox et al. 2022 ). Similarly, production planning , a longstanding research area, has begun to harness sensor-driven data to improve production schedules, processes, quality, and operations. Examples include PAS to optimize shop floor operations (Stein et al. 2018 ) and schedule diffusion furnaces (Vimala Rani and Mathirajan 2021 ).

In addition to the two dominant affordances, there are other, less explored examples within the manufacturing sector. Some of these include supporting product development , optimizing product portfolio designs (Jank et al. 2019 ), and enhancing the design of industrial products (Dey et al. 2019 ). Moreover, research has proposed utilizing PAS for training industrial workers by offering training schedules based on digital twins (Longo et al. 2023 ), prescribing optimal safety actions (Soroush et al. 2020 ), and improving additive manufacturing processes through deformation control (Jin et al. 2016 ). As the manufacturing industry continues to advance, the potential applications of PAS are expected to grow, fostering further innovation and efficiency.

4.3.2 Transportation and logistics

Facing significant pressure regarding cost efficiency and sustainability, the transportation and logistics industry, like manufacturing, has a history of using mathematical optimization models for routing and scheduling tasks (Konstantakopoulos et al. 2022 ). With vehicles becoming increasingly connected and generating digital traces through sensors and networks, PAS can further harness this data to afford efficiency, enabling the industry to capitalize on various applications. Routing and scheduling naturally emerge as the most researched affordance.

Examples span various modes of transportation, such as ground (Gutierrez-Franco et al. 2021 ), air (Ayhan et al. 2018 ), and public transport (Xylia et al. 2016 ), showcasing the versatility and potential of PAS in enhancing efficiency across diverse systems. Another affordance is capacity management . Affordance effects include optimizing freight or cargo distribution (Rizzo et al. 2020 ) and passenger seat assignments (Moore et al. 2021 ), ensuring efficient resource allocation, and enhancing overall operational performance. Lastly, considering that vehicles undergo continuous degradation while in use, our sample also includes prescriptive maintenance affordances to effectively address wear and tear, optimize maintenance schedules, and prolong the service life of vehicles (Consilvio et al. 2019 ; Anglou et al. 2021 ).

4.3.3 Health and MedTech

Health and MedTech are the second most researched sectors in our literature sample. Implementing PAS can significantly improve resource allocation and overall patient outcomes with the increasing complexity and demand for healthcare services. The sector features two dominant affordances: patient treatment planning and scheduling. Patient treatment planning primarily focuses on improving health outcomes by optimizing treatments, reducing hospital readmissions, enhancing precision medicine, and boosting clinical staff efficiency (Rider et al. 2021 ; Zheng et al. 2021 ). On the other hand, patient scheduling emphasizes the efficient management of appointment scheduling and bed occupancy (Belciug and Gorunescu 2016 ; Srinivas and Ravindran 2018 ). In the clinical context, niche examples of affordances include assortment, inventory planning (Galli et al. 2021 ) , and investment management (Fang et al. 2021 ).

Further, in light of the recent COVID-19 pandemic, several contributions have focused on pandemic or epidemic intervention planning . These studies consider various aspects, such as mobility intervention and the rapid deployment of medical staff and equipment (Miikkulainen et al. 2021 ; Ahmed et al. 2021 ). Lastly, a group of researchers has shifted their focus to the patients or their bodies directly, for example, incorporating sensor data for health tracking , enhancing safety and consumption decisions, and preventing impulsive behavior among patients (Sedighi Maman et al. 2020 ; Raychaudhuri et al. 2021 ). By leveraging PAS in these areas, healthcare providers can offer more personalized care, empower patients to make better-informed decisions, and ultimately improve overall health outcomes.

4.3.4 Energy and environment

The energy and environment industry showcases more diverse affordances in our literature sample than in previous sectors. A key focus in this domain is power generation systems (e.g., wind farms), where PAS are utilized to afford optimized performance (Tektaş et al. 2022 ), electricity brokerage (Peters et al. 2013 ), and prescriptive maintenance (Goyal et al. 2016 ) for these systems. Additionally, PAS applications extend to optimizing waste collection and planning (Vargas et al. 2022 ) and enhancing wastewater treatment processes (Zadorojniy et al. 2019 ). Some PAS afford disaster preparation and recovery planning , addressing challenges posed by wildfires, hurricanes, or floods (Hu et al. 2019 ; Yang et al. 2022 ). Niche examples within this sector include soil slope analysis (Li et al. 2019 ) and optimization of battery lifetime (Eider and Berl 2020 ). As demonstrated for this sector, implementing AI systems can significantly impact society by improving sustainability, efficiency, and environmental protection (Schoormann et al. 2023 ).

4.3.5 Retail and trade

The retail and trade industry encompasses both B2B and B2C interactions. Despite its size, there has been relatively little PAS research in this area compared to other industries. One possible reason may be the traditional set-up often found in brick-and-mortar stores and a comparatively lower level of digitization than in industries like manufacturing or logistics. Nevertheless, there are instances of PAS research in our sample. One example is dynamic price optimization , which incorporates factors such as customers, competition, business partners, and environmental aspects (Ito and Fujimaki 2017 ). A key challenge in this industry is assortment and inventory planning , which is a significant cost driver when considering perishable goods or inventory costs (Jin et al. 2016 ; Flamand et al. 2018 ). In the B2B context, sales teams drive revenue, making optimal assignment a potential affordance. Other examples include customer characterization (Perugini and Perugini 2014 ), customer service recommendations (Lo and Pachamanova 2015 ), theft surveillance, and facilitating automated store checkouts (Hauser et al. 2021 ).

4.3.6 Education

Three primary affordances have emerged in education: dropout prevention planning, improving students’ academic performance, and admissions planning. Firstly, dropout prevention planning focuses on identifying students at risk of leaving their educational programs prematurely. PAS enables institutions to target support and interventions, ensuring that students receive the help they need to stay on track and complete their studies (Yanta et al. 2021 ; de Jesus and Ledda 2021 ). Secondly, improving academic performance is an essential priority for educational institutions. By utilizing PAS, educators can gain insights into students’ learning patterns and areas of difficulty, enabling them to tailor teaching approaches and offer personalized learning experiences that foster success (Uskov et al. 2019 ; Islam et al. 2021 ). Finally, admissions planning is essential to maintaining a thriving educational institution. PAS can help optimize the admissions process by analyzing student demographics and academic performance to ensure that institutions admit the most suitable candidates (Kiaghadi and Hoseinpour 2023 ). Applying PAS in education can improve traditional academic decision-making processes, enhance student outcomes, and streamline institutional operations.

4.3.7 Chemicals and resources

The chemicals and resources industry has some affordances identified in our review, aiming to optimize processes, boost efficiency, and promote sustainable resource utilization. Maximizing oil and gas recovery is the most researched affordance effect, with PAS used to optimize extraction techniques, reservoir modeling, and resource management. Other examples include laboratory task allocation (Silva and Cortez 2022 ), mining fleet scheduling (Nakousi et al. 2018 ), optimizing biodiesel properties (Suvarna et al. 2022 ) , and improving sand molding processes (Chowdhary and Khandelwal 2018 ).

4.3.8 Technology and communication

Despite its technological maturity, the technology and communications sector has limited research on PAS applications. This sector can benefit from PAS in various affordances, such as network and computing resource orchestration (Ceselli et al. 2018 , 2019 ), social media optimization (Ballings et al. 2016 ), software development estimation (Pospieszny 2017 ), and website performance analysis (Salvio and Palaoag 2019 ). PAS can streamline network and computing systems, enhance social media campaigns, improve project management, and optimize website performance. Although current literature is limited, this industry has the potential for further exploration. Harnessing PAS can improve performance, efficiency, and user satisfaction across technology and communication operations.

4.3.9 Academia

Furthermore, several studies have investigated the use of PAS in academia. They focus on improving and enhancing academic research performance. By analyzing data related to research output and other relevant factors, such as citations or related work, PAS can provide valuable insights and recommendations for researchers, such as journals or references. Additionally, some research has explored the concept of system thinking and crafting scenarios. These approaches help researchers better understand their work’s potential outcomes and consequences, enabling them to make more informed decisions about their research directions (e.g., Song et al. 2014 ; Jeong and Joo 2019 ).

4.3.10 Industry-agnostic

Beyond sector-specific applications, much research investigates affordances from an industry-agnostic perspective. In these cases, authors often apply prescriptive analytics in a context, which we coin prescriptive process management (Krumeich et al. 2016 ; Kubrak et al. 2022 ). The primary affordance effects include process monitoring, controlling execution, and recommending the most appropriate subsequent actions. Interestingly, two contributions specifically discuss the explainability of decision outputs in this setting (Mehdiyev and Fettke 2020 ; Notz 2020 ). Moreover, PAS can be used to optimize employee recruitment (Pessach et al. 2020 ), facility and asset management (Lavy et al. 2014 ), and supplier selection (Abdollahnejadbarough et al. 2020 ). Some niche systems involve enhancing accounting procedures and improving call center routing (Ali 2011 ). Overall, these versatile applications demonstrate the potential of PAS to streamline operations and support decision-making across a wide range of industry contexts.

4.3.11 Summary

In conclusion, PAS have demonstrated the potential to affect various industries positively, for instance, by optimizing processes, allocating resources, scheduling, and planning maintenance actions. The systems afford organizations to achieve enhanced efficiency and productivity. Although the current research varies in scope and depth across different sectors, the widespread applicability of PAS indicates its capacity to drive innovation and streamline operations across a wide range of contexts. It is important to note that this section did not detail all affordances, specifically underrepresented industries such as tourism, media, and finance. Please refer to Table  6 and the example papers for more information on these sectors and their respective affordances.

Even though the PAS in our literature sample are diverse, they share the commonality of prescribing the best course of action in a specific decision environment and situation. Given the abovementioned affordances, we derived three overarching affordance effects of PAS to improve decision-making processes: (1) improvement, (2) scheduling, and (3) resource allocation, which we detail in the following.

Improvement (1) is centered on enhancing and optimizing the current state of an object or the decision environment to achieve an improved state. This affordance involves conducting a comprehensive analysis and adjusting processes, products, or operations to make them more effective, efficient, and aligned with predefined goals. For example, product design optimization focuses on continuously refining products based on user feedback and market trends (e.g., Dey et al. 2019 ; Jank et al. 2019 ). Likewise, within the healthcare sector, patient treatment planning and improvement entail tailoring treatments to individual patient needs while constantly updating these plans based on patient responses and emerging medical insights (e.g., Rider et al. 2021 ; Zheng et al. 2021 ). Similarly, maintenance optimization aims to optimize equipment performance and longevity in industrial settings, ultimately reducing downtime and increasing efficiency (e.g., Liu et al. 2019 ; Consilvio et al. 2019 ) .

Scheduling (2) entails strategically organizing and coordinating tasks and actions in the decision environment over time, creating and managing a timeline of activities that aligns with an organization’s objectives, resource availability, and external factors. Effective scheduling and planning improve operational flow and resource utilization, reducing bottlenecks and inefficiencies. For example, in healthcare, patient scheduling is critical for maximizing medical facilities and staff use (e.g., Belciug and Gorunescu 2016 ; Srinivas and Ravindran 2018 ), while in the energy sector, planning energy distribution is crucial for balancing supply with consumer demand (Goyal et al. 2016 ).

Resource allocation (3) involves strategically distributing resources like workforce, materials, and finances to areas most needed and will be most effective. This process requires a thorough understanding of resources’ availability, potential, limitations, and different organizations’ objectives and needs. For example, in logistics, capacity and cargo management ensures optimal use of transport and storage resources (Rizzo et al. 2020 ; Gutierrez-Franco et al. 2021 ). In urban planning, efficient allocation of resources for waste collection and management is vital for maintaining cleanliness and public health (Vargas et al. 2022 ).

The affordances are not mutually exclusive but can overlap depending on the specific system, application use case, or complex organizational settings. For instance, in manufacturing, the improvement of a product design (Improvement) is closely linked to the planning of production schedules (Scheduling) and the allocation of manufacturing resources (Resource Allocation). Similarly, in healthcare, patient treatment plans (Improvement) need to be integrated with patient scheduling (Scheduling and Planning) and the allocation of medical staff and equipment (Resource Allocation). Understanding the interplay and overlap of these affordances is crucial for effective management and decision-making. Organizations can holistically approach problem-solving and optimization by recognizing how they complement each other, leading to more comprehensive PAS.

5 Directions for future research

Drawing on the synthesis and conceptualization in the previous sections, we discuss the main observations and deriving aspects that remain open to pave the ground for future research. We will discuss possible research directions from technical, social, and overarching perspectives. Table 7 highlights the research agenda, key observations, and illustrative research directions or questions.

5.1 Technical perspectives

The preponderance of research within the technical subsystem is understandable, as PAS are fundamentally rooted in technology, and a significant portion of investigations in this area stem from disciplines closely tied to technological advancements. Our SLR has uncovered numerous PAS components that various authors in our sample have extensively researched and well-addressed, establishing a solid foundation in the field. Despite this wealth of knowledge, we have identified specific gaps that warrant further investigation.

For instance, action mechanisms have been relatively underexplored in the existing literature. We contend that these components are crucial for designing an effective PAS, as they drive the interaction between the prescriptive agent and the human decision-maker. Further, ancillary features with a focus on seamlessly integrating technology components within the social structure and the broader organizational landscape are lacking in current research.

In addition to the identified gaps, mathematical programming and, to a lesser extent, bio-inspired optimization algorithms have been well-established in prescriptive analytics. Recently, there has been a surge of interest in incorporating ML techniques to integrate predictions or probabilities as precursors to subsequent optimization processes. ML, particularly deep learning, is widely researched for managing extensive and high-dimensional datasets (Janiesch et al. 2021 ). However, challenges surrounding interpretability and explainability have hindered its adoption among decision-makers. While explainable and interpretable ML offer promising solutions (Zschech et al. 2022 ; Herm et al. 2022 ; Wanner et al. 2022 ), integrating these approaches into a PAS remains mainly open. Today, ML predominantly contributes to descriptive and predictive analytics, but increased transparency and trust, which are vital, especially for high-stakes decision-making, remain open. However, interpretable ML would be a valuable extension to existing prescriptive analytics approaches (Shollo et al. 2022 ).

Further, RL has also garnered attention for its potential in PAS owing to its dynamic, adaptive, and iterative nature and its aptitude for addressing well-formalized decision-making problems (Greene et al. 2022 ). Researchers have already demonstrated the effectiveness of RL in the context of adaptive and self-governing PAS archetypes. However, our understanding is limited by the scarcity of literature on this topic. Consequently, further research is needed to explore RL’s unique characteristics, concepts, and requirements within the PAS framework, ultimately contributing to a more robust understanding of its potential applications and benefits and an alternative to more traditional optimization techniques such as linear programming.

Finally, As a recent innovation in AI, foundation models – particularly large language models – have catalyzed a significant paradigm shift in the development of AI systems. This transformation has already profoundly impacted existing IT services and ecosystems while simultaneously enabling the creation of novel applications (Feuerriegel et al. 2024 ; Schneider et al. 2024 ). Through our review, we have observed that the application of foundation models in prescriptive analytics remains unexplored. However, we posit that this domain holds substantial potential. Future research should thus investigate leveraging the advanced capabilities of foundation models to enhance intelligent decision-making systems.

5.2 Social perspectives

As previously mentioned, the social subsystem (including the human decision-maker) has been relatively underresearched, which is somewhat understandable considering the technological origins of prescriptive analytics. However, we argue that increased attention to this perspective is essential for a more comprehensive understanding of the field, especially for the IS community.

Current research in this area often takes a case-specific approach, with insufficient consideration of the broader organizational landscape and how systems integrate into the larger picture. The outputs generated by these decision-making processes can have varying contextual implications, depending on factors such as the business unit or hierarchical level, ultimately influencing strategic and operational decision-making (e.g., Appelbaum et al. 2017 ). Consequently, we posit that a PAS will be most valuable if it is organization-wide, encompassing all decision processes and business functions, avoiding siloed structures, and made available (e.g., as-a-service) to all decision-makers tailored to their unique environments. From a technological standpoint, achieving this vision requires infrastructural features such as standardized data integration, interoperability, distributed computing, and effective API design (Lepenioti et al. 2020 ; Vieira et al. 2020 ; Verbraeken et al. 2021 ). However, further research is needed to identify the specific requirements unique to a PAS. Additionally, given that analytics is not solely a technology-driven concept and necessitates a cultural shift towards evidence-based decision-making, it is worth exploring whether a PAS should be a mere result of this shift or function as a driver or initiative to move organizations toward more fact-based decision processes.

Beyond the organizational setting, the human decision-maker’s perspective warrants further research. Specifically, ML for predictive analytics has been extensively studied regarding explainability, interpretability, accountability, fairness, and bias (e.g., Meske et al. 2022 ; Nadeem et al. 2022 ; Kraus et al. 2023 ). Due to the inherent differences between various analytics methods and outputs, particularly those focused on prescriptive analytics, we argue that additional research is needed, specifically in the context of PAS, such as the decision-maker’s trust in decision outputs (Caro and de Tejada Cuenca 2023 ).

This line of inquiry will facilitate a deeper understanding of the challenges and opportunities associated with PAS. It will also foster their responsible development and deployment within organizations, ensuring that they align with ethical standards and contribute positively to decision-making processes.

5.3 General perspectives

Our synthesis identified four archetypes within PAS. Despite this progress, most PAS in our sample are predominantly advisory, while the executive, adaptive, and self-governing archetypes remain underexplored. This imbalance suggests a significant disconnect between problem environments and PAS, as these systems are often disjointed from the last two phases of the decision-making process. In real-world scenarios, many environments are constantly in flux due to natural changes or actions. This dynamic nature is frequently overlooked in current PAS research. However, action execution, adaptation, and learning mechanisms hold great potential, as they can help reduce information loss across iterations and improve decision-making processes over time while minimizing reliance on subjective or judgmental human experiences (Sturm et al. 2021 ). The BA community must address this gap by understanding the requirements of such systems and developing case-agnostic blueprints with corresponding design principles and options.

Furthermore, delegation mechanisms warrant increased attention, representing the initial steps toward hybrid intelligence systems and the symbiosis between agents and humans in decision-making (Dellermann et al. 2019 ; Peng et al. 2022 ). By focusing on these underexplored areas, researchers can contribute to a more comprehensive understanding of PAS, ultimately fostering the development of systems that effectively integrate advanced technology and human expertise in organizational decision-making processes.

In our review, we utilized the affordance theory to examine how individuals or organizations with specific objectives can leverage technology at a basic level. Given the nature of the papers in our sample, we focused on affordance effects. We argue that understanding the balance between technology’s enabling and constraining aspects is crucial for designing effective PAS for organizational decision-making. Affordance theory offers a promising perspective for future research in analytics-driven organizational decision-making. One area to explore is the process of affordance-actualization, which involves understanding how the potential benefits of technology are actualized in the form of organizational outcomes (Strong et al. 2014 ). In the context of our work, this would involve investigating how decision outputs from a PAS are implemented within organizations and the effects on decision-making processes themselves.

Another direction is applying the affordance-network approach, which examines how organizations achieve more significant outcomes by connecting a series of more immediate, concrete decision outcomes within a network of interrelated affordances (Burton-Jones and Volkoff 2017 ). Furthermore, exploring the trajectory along which affordances travel can provide valuable insights into the processes and conditions that shape the perception and actualization of affordances in organizational decision-making (Thapa and Sein 2018 ). This line of inquiry can help elucidate how the potential benefits of a PAS are transformed into tangible outcomes in practice. By considering these aspects of affordance theory, researchers and practitioners could develop more effective and comprehensive PAS for decision-making in organizational settings.

6 Concluding remarks

Our research contributes to the knowledge of BA, specifically PAS, as the most sophisticated maturity level to support organizational decision-making with the highest potential business value. In this context, we conducted an SLR on the state of PAS research, emphasizing the IS artifact and GST as a theoretical framework. We reviewed 262 relevant contributions, enabling a holistic view of the field and its relevance to the broader BA community in research and practice, with three main contributions.

Our first contribution is the development of a concept matrix comprising 23 distinct constituent components, revealing fundamental technical design elements. Based on GST and considering important ancillary features, this updated understanding gives researchers a starting point to study the relationships between elements and their effective integration into organizational decision processes.

Second, we analyzed the meta-level of PAS, revealing the differing roles of prescriptive agents and human decision-makers from a decision-theoretic perspective and highlighting their synergy, delegation, and adaptability. Our conceptualization led us to derive four archetypes, each providing varying levels of support in the decision-making process. From a practical perspective, our findings can serve as initial blueprints or guiding principles on what system modules and features to consider when designing a PAS for specific purposes.

Third, we identified various technology affordance effects of PAS across various sectors, unveiling the purpose and benefits of employing such systems. Owing to the extensive chronology of leveraging optimization algorithms within industries like manufacturing and transportation, a data-driven approach to prescriptive analytics has experienced considerable exploration in these sectors. Meanwhile, although specific fields exhibit a reduced degree of investigation, a comprehensive assessment revealed a heterogeneous array of research endeavors spanning multiple industries. Finally, our paper reveals six key findings or observations, enabling us to derive various research directions and implications for the BA and IS community.

As with most research endeavors, our study comes with certain limitations. Our goal was to establish the status quo and create a shared understanding of the existing body of knowledge on PAS. To achieve this, we relied on concept matrices as a qualitative analysis tool, which is never complete and serves as an initial foundation for more comprehensive research and contextualization. Furthermore, we focus our SLR on prescriptive analytics, resulting in an underrepresentation of the socio-technical lens in our literature sample due to the current research’s technological and algorithmic focus. However, even though the papers did not explicitly mention the social subsystem or interactions, we could make educated assumptions based on, for example, architectural overviews presented, which enabled us to derive system archetypes, such as how different components synergize with the decision-maker.

Additionally, it is imperative to highlight that our primary objective was to demystify the PAS landscape in research from an overarching IS perspective. We endeavored to conceptualize and categorize the aspects within this domain, acknowledging the inherent challenges in attributing specific concepts to distinct groups, especially for prescriptive analytics techniques. While we recognize that the clarity of separation between these elements is not always absolute, owing to the multifaceted nature and numerous influencing factors, we believe our work provides a substantial foundation. This framework is particularly beneficial for newcomers in the field, whether designing systems or conducting research. It offers a solid grounding, enabling a deeper and more nuanced understanding of the PAS space. We assert that this is essential for anyone aspiring to navigate, contribute to, or innovate within this ever-evolving and complex field. However, additional relevant PAS perspectives may still be explored, for example, within a deeper algorithmic review scope. For instance, a potential area of investigation could be the differentiation between sequential (predict-then-optimize) and simultaneous (predict-and-optimize) PAS. Here, the latter could potentially lead to improved decision outcomes, as it proposes learning a predictive model by directly minimizing the cost of the downstream decision-making task (Vanderschueren et al. 2022 ; Zhang et al. 2022 ). Taking this as an example, future reviews with an algorithmic focus would be valuable avenues of inquiry, especially given the recent AI model innovations, which may significantly impact how future PAS are designed.

Further, we note that "prescription" or "prescriptive analytics" may not be used as frequently in every research discipline. Some contributions may only refer to an affordance (e.g., scheduling, routing) or the technique (e.g., optimization, linear programming) in their title, keyword, or abstract, possibly excluding these with our search string. Although prescriptive analytics originated in the BA domain (Holsapple et al. 2014 ; Delen and Zolbanin 2018 ), our review revealed that it is a well-established concept in various research disciplines today, with special issues dedicated to the topic in widely respected journals (e.g., Giesecke et al. 2022 ). It is clear that today, prescriptive analytics is an overarching concept, describing a task objective or even a decision-making paradigm nested in a socio-technical system instead of a specific technique or algorithm to be employed, such as linear programming or a specific ML technique. This trend is also reflected in our sample’s growing number of publications after 2018, enabling us to establish a representative view of the current research state with many contemporary examples from the literature. However, our literature-centric approach must be assessed for practical relevance by incorporating real-world PAS applications and engaging with practitioners with hands-on experience in the field. This perspective will ensure a well-rounded understanding of the topic and its implications and help bridge the gap between theory and practice.

In summary, our research provides a comprehensive understanding of the current state of PAS and highlights areas for future research and development. By exploring these opportunities, researchers and practitioners can collaborate to create more effective and efficient PAS, ultimately driving better decision-making and business value in organizations.

Abdollahnejadbarough H, Mupparaju KS, Shah S, Golding CP, Leites AC, Popp TD, Shroyer E, Golany YS, Robinson AG, Akgun V (2020) Verizon uses advanced analytics to rationalize its tail spend suppliers. INFORMS J on Appl Anal 50(3):197–211. https://doi.org/10.1287/inte.2020.1038

Article   Google Scholar  

Ahmed I, Ahmad M, Jeon G, Piccialli F (2021) A Framework for pandemic prediction using big data analytics. Big Data Res 25:100190. https://doi.org/10.1016/j.bdr.2021.100190

Aier S, Fischer C (2011) Criteria of progress for information systems design theories. Inf Syst E-Bus Manage 9(1):133–172. https://doi.org/10.1007/s10257-010-0130-8

Ali AR (2011) Intelligent call routing: optimizing contact center throughput. In: Proceedings of the Eleventh International Workshop on Multimedia Data Mining

Anderson C, Robey D (2017) Affordance potency: explaining the actualization of technology affordances. Inf Organ 27(2):100–115. https://doi.org/10.1016/j.infoandorg.2017.03.002

Anglou F-Z, Ponis S, Spanos A (2021) A machine learning approach to enable bulk orders of critical spare-parts in the shipping industry. J Ind Eng 14(3):604. https://doi.org/10.3926/jiem.3446

Ansari F, Glawar R, Nemeth T (2019) PriMa: a prescriptive maintenance model for cyber-physical production systems. Int J Comput Integr Manuf 32(4–5):482–503. https://doi.org/10.1080/0951192X.2019.1571236

Appelbaum D, Kogan A, Vasarhelyi M, Yan Z (2017) Impact of business analytics and enterprise systems on managerial accounting. Int J of Account Inf Syst 25:29–44. https://doi.org/10.1016/j.accinf.2017.03.003

Ayhan S, Costas P, Samet H (2018) Prescriptive analytics system for long-range aircraft conflict detection and resolution. In: Proceedings of the 26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. pp 239–248

Baird A, Maruping LM (2021) The next generation of research on is use: a theoretical framework of delegation to and from agentic IS artifacts. MISQ 45(1):315–341. https://doi.org/10.25300/MISQ/2021/15882

Ballings M, Van den Poel D, Bogaert M (2016) Social media optimization: identifying an optimal strategy for increasing network size on Facebook. Omega 59:15–25. https://doi.org/10.1016/j.omega.2015.04.017

Basdere M, Caniglia G, Collar C, Rozolis C, Chiampas G, Nishi M, Smilowitz K (2019) SAFE: a comprehensive data visualization system. INFORMS J on Appl Anal 49(4):249–261. https://doi.org/10.1287/inte.2019.0989

Belciug S, Gorunescu F (2016) A hybrid genetic algorithm-queuing multi-compartment model for optimizing inpatient bed occupancy and associated costs. Artif Intell in Med 68:59–69. https://doi.org/10.1016/j.artmed.2016.03.001

Bertsimas D, Kallus N (2020) From predictive to prescriptive analytics. Manage Sci 66(3):1025–1044. https://doi.org/10.1287/mnsc.2018.3253

Bhatt D, Naqvi S, Gunasekaran A, Dutta V (2023) Prescriptive analytics applications in sustainable operations research: conceptual framework and future research challenges. Ann Oper Res. https://doi.org/10.1007/s10479-023-05251-3

Bostrom RP, Heinen JS (1977) MIS Problems and Failures: a socio-technical perspective, part II: the application of socio-technical theory. MIS Q 1(4):11. https://doi.org/10.2307/249019

Bousdekis A, Papageorgiou N, Magoutas B, Apostolou D, Mentzas G (2020) Sensor-driven learning of time-dependent parameters for prescriptive analytics. IEEE Access 8:92383–92392. https://doi.org/10.1109/ACCESS.2020.2994933

Brandt T, Dlugosch O, Abdelwahed A, van den Berg PL, Neumann D (2021) Prescriptive analytics in urban policing operations. Manuf Serv Oper Manage 24(5):2463–2480. https://doi.org/10.1287/msom.2021.1022

vom Brocke J, Simons A, Riemer K, Niehaves B, Plattfaut R, Cleven A (2015) Standing on the shoulders of giants: challenges and recommendations of literature search in information systems research. Commun Assoc Inf Syst 37:206–224. https://doi.org/10.17705/1CAIS.03709

Burton-Jones A, Volkoff O (2017) How can we develop contextualized theories of effective use? a demonstration in the context of community-care electronic health records. Inf Syst Res 28(3):468–489. https://doi.org/10.1287/isre.2017.0702

Caro F, de Tejada Cuenca AS (2023) Believing in analytics: managers’ adherence to price recommendations from a DSS. M&SOM 25(2):524–542. https://doi.org/10.1287/msom.2022.1166

Ceselli A, Fiore M, Premoli M, Secci S (2019) Optimized assignment patterns in mobile edge cloud networks. Comput Oper Res 106:246–259. https://doi.org/10.1016/j.cor.2018.02.022

Ceselli A, Fiore M, Furno A, Premoli M, Secci S, Stanica R (2018) Prescriptive Analytics for MEC Orchestration. In: 2018 IFIP Networking Conference (IFIP Networking) and Workshops. IEEE, Zurich, Switzerland, pp 1–9

Chatterjee S, Sarker S, Lee MJ, Xiao X, Elbanna A (2021) A possible conceptualization of the information systems (IS) artifact: A general systems theory perspective. Inf Syst J 31(4):550–578. https://doi.org/10.1111/isj.12320

Chowdhary D, Khandelwal H (2018) Data Analytics: The next dimension in molding sand control. In: The 73rd World Foundry Congress

Consilvio A, Sanetti P, Anguita D, Crovetto C, Dambra C, Oneto L, Papa F, Sacco N (2019) Prescriptive Maintenance of Railway Infrastructure: From Data Analytics to Decision Support. In: 2019 6th International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS). IEEE, Cracow, Poland, pp 1–10

Cooper HM (1988) Organizing knowledge syntheses: a taxonomy of literature reviews. Knowl in Soc 1(1):104–126. https://doi.org/10.1007/BF03177550

Darioshi R, Lahav E (2021) The impact of technology on the human decision-making process. Human Behav and Emerg Tech 3(3):391–400. https://doi.org/10.1002/hbe2.257

de Jesus CG, Ledda MKC (2021) Intervention Support Program for Students at Risk of Dropping Out Using Fuzzy Logic-Based Prescriptive Analytics. In: 2021 IEEE 17th International Colloquium on Signal Processing & Its Applications (CSPA). IEEE, Langkawi, Malaysia, pp 144–149

Delen D, Ram S (2018) Research challenges and opportunities in business analytics. J Bus Anal 1(1):2–12. https://doi.org/10.1080/2573234X.2018.1507324

Delen D, Zolbanin HM (2018) The analytics paradigm in business research. J Bus Res 90:186–195. https://doi.org/10.1016/j.jbusres.2018.05.013

Dellermann D, Ebel P, Söllner M, Leimeister JM (2019) Hybrid intelligence. Bus Inf Syst Eng 61(5):637–643. https://doi.org/10.1007/s12599-019-00595-2

Dey S, Gupta N, Pathak S, Kela DH, Datta S (2019) Data-Driven Design Optimization for Industrial Products. In: Datta S, Davim JP (eds) Optimization in Industry. Springer International Publishing, Cham, pp 253–267

Chapter   Google Scholar  

Effah J, Amankwah-Sarfo F, Boateng R (2021) Affordances and constraints processes of smart service systems: insights from the case of seaport security in Ghana. Int J Inf Manage 58:102204. https://doi.org/10.1016/j.ijinfomgt.2020.102204

Eider M, Berl A (2020) Requirements for Prescriptive Recommender Systems Extending the Lifetime of EV Batteries. In: 2020 10th International Conference on Advanced Computer Information Technologies (ACIT). IEEE, Deggendorf, Germany, pp 412–417

Elmachtoub AN, Grigas P (2022) Smart “predict, then optimize.” Manage Sci 68(1):9–26. https://doi.org/10.1287/mnsc.2020.3922

Fang X, Gao Y, Jen-Hwa HuP (2021) A prescriptive analytics method for cost reduction in clinical decision making. MIS Q 45(1):83–115. https://doi.org/10.25300/MISQ/2021/14372

Feuerriegel S, Hartmann J, Janiesch C, Zschech P (2024) Generative AI. Bus Inf Syst Eng 66(1):111–126. https://doi.org/10.1007/s12599-023-00834-7

Flamand T, Ghoniem A, Haouari M, Maddah B (2018) Integrated assortment planning and store-wide shelf space allocation: an optimization-based approach. Omega 81:134–149. https://doi.org/10.1016/j.omega.2017.10.006

Fox H, Pillai AC, Friedrich D, Collu M, Dawood T, Johanning L (2022) A review of predictive and prescriptive offshore wind farm operation and maintenance. Energies 15(2):504. https://doi.org/10.3390/en15020504

Frazzetto D, Nielsen TD, Pedersen TB, Šikšnys L (2019) Prescriptive analytics: a survey of emerging trends and technologies. VLDB J 28(4):575–595. https://doi.org/10.1007/s00778-019-00539-y

Galli L, Levato T, Schoen F, Tigli L (2021) Prescriptive analytics for inventory management in health care. J Oper Res Soc 72(10):2211–2224. https://doi.org/10.1080/01605682.2020.1776167

Gibson JJ (1986) The Ecological Approach to Visual Perception. Lawrence Erlbaum Asso, Hillsdale, NJ

Giesecke K, Liberali G, Nazerzadeh H, Shanthikumar JG, Teo CP (2022) Introduction to the special section on data-driven prescriptive analytics. Manage Sci 68(3):1591–1594. https://doi.org/10.1287/mnsc.2021.4296

Gordon CAK, Burnak B, Onel M, Pistikopoulos EN (2020) Data-driven prescriptive maintenance: failure prediction using ensemble support vector classification for optimal process and maintenance scheduling. Ind Eng Chem Res 59(44):19607–19622. https://doi.org/10.1021/acs.iecr.0c03241

Goyal A, Aprilia E, Janssen G, Kim Y, Kumar T, Mueller R, Phan D, Raman A, Schuddebeurs J, Xiong J, Zhang R (2016) Asset health management using predictive and prescriptive analytics for the electric power grid. IBM J Res Dev 60(1):4:1-4:14

Greene T, Shmueli G, Ray S (2022) Taking the Person Seriously: Ethically-aware IS Research in the Era of Reinforcement Learning-based Personalization. J Assoc Inf 77:Available at: https://aisel.aisnet.org/jais_preprints/77

Gutierrez-Franco E, Mejia-Argueta C, Rabelo L (2021) Data-driven methodology to support long-lasting logistics and decision making for urban last-mile operations. Sustainability 13(11):6230. https://doi.org/10.3390/su13116230

Harikumar H, Rana S, Gupta S, Nguyen T, Kaimal R, Venkatesh S (2022) Prescriptive analytics with differential privacy. Int J Data Sci Anal 13(2):123–138. https://doi.org/10.1007/s41060-021-00286-w

Hauser M, Flath CM, Thiesse F (2021) Catch me if you scan: data-driven prescriptive modeling for smart store environments. Eur J Oper Res 294(3):860–873. https://doi.org/10.1016/j.ejor.2020.12.047

Herm L-V, Steinbach T, Wanner J, Janiesch C (2022) A nascent design theory for explainable intelligent systems. Electron Mark 32(4):2185–2205. https://doi.org/10.1007/s12525-022-00606-3

Hinsen S, Hofmann P, Jöhnk J, Urbach N (2022) How Can Organizations Design Purposeful Human-AI Interactions: A Practical Perspective From Existing Use Cases and Interviews. In: Hawaii International Conference on System Sciences (HICSS)

Holsapple C, Lee-Post A, Pakath R (2014) A unified foundation for business analytics. Decis Support Syst 64:130–141. https://doi.org/10.1016/j.dss.2014.05.013

Hu R, Fang F, Pain CC, Navon IM (2019) Rapid spatio-temporal flood prediction and uncertainty quantification using a deep learning method. J Hydrol 575:911–920. https://doi.org/10.1016/j.jhydrol.2019.05.087

Islam S, Mouratidis H, Mahmud H (2021) An Automated Tool to Support an Intelligence Learner Management System Using Learning Analytics and Machine Learning. In: Maglogiannis I, Macintyre J, Iliadis L (eds) Artificial intelligence applications and innovations. Springer International Publishing, Cham, pp 494–504

Ito S, Fujimaki R (2017) Optimization Beyond Prediction: Prescriptive Price Optimization. In: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, Halifax NS Canada, pp 1833–1841

Janiesch C, Zschech P, Heinrich K (2021) Machine learning and deep learning. Electron Mark 31(3):685–695. https://doi.org/10.1007/s12525-021-00475-2

Jank M-H, Dölle C, Schuh G (2019) Product Portfolio Design Using Prescriptive Analytics. In: Schmitt R, Schuh G (eds) Advances in production research. Springer International Publishing, Cham, pp 584–593

Jensen MH, Persson JS, Nielsen PA (2023) Measuring benefits from big data analytics projects: an action research study. Inf Syst E-Bus Manage 21(2):323–352. https://doi.org/10.1007/s10257-022-00620-0

Jeong D-H, Joo H-S (2019) Topical prescriptive analytics system for automatic recommendation of convergence technology. Biotechnol Bioprocess Eng 24(6):893–906. https://doi.org/10.1007/s12257-019-0305-1

Jin Y, Qin SJ, Huang Q (2016) Prescriptive analytics for understanding of out-of-plane deformation in additive manufacturing. In: 2016 IEEE International Conference on Automation Science and Engineering (CASE). IEEE, Fort Worth, TX, USA, pp 786–791

Jozefowiez N, Semet F, Talbi E-G (2008) Multi-objective vehicle routing problems. Eur J Oper Res 189(2):293–309. https://doi.org/10.1016/j.ejor.2007.05.055

Kast FE, Rosenzweig JE (1972) General system theory: applications for organization and management. Acad Manage J 15(4):447–465. https://doi.org/10.2307/255141

Kawas B, Squillante MS, Subramanian D, Varshney KR (2013) Prescriptive Analytics for Allocating Sales Teams to Opportunities. In: IEEE 13th International Conference on Data Mining Workshops. IEEE, pp 211–218

Kiaghadi M, Hoseinpour P (2023) University admission process: a prescriptive analytics approach. Artif Intell Rev 56(1):233–256. https://doi.org/10.1007/s10462-022-10171-y

Kim J-S, Jin H, Züfle A (2020) Expert-in-the-Loop Prescriptive Analytics using Mobility Intervention for Epidemics. In: International Workshop on Prescriptive Analytics for the Physical World

Knabke T, Olbrich S (2018) Building novel capabilities to enable business intelligence agility: results from a quantitative study. Inf Syst E-Bus Manage 16(3):493–546. https://doi.org/10.1007/s10257-017-0361-z

Konstantakopoulos GD, Gayialis SP, Kechagias EP (2022) Vehicle routing problem and related algorithms for logistics distribution: a literature review and classification. Oper Res Int J 22(3):2033–2062. https://doi.org/10.1007/s12351-020-00600-7

Kraus M, Tschernutter D, Weinzierl S, Zschech P (2023) Interpretable generalized additive neural networks. Eur J Oper Res. https://doi.org/10.1016/j.ejor.2023.06.032

Krumeich J, Werth D, Loos P (2016) Prescriptive control of business processes: new potentials through predictive analytics of big data in the process manufacturing industry. Bus Inf Syst Eng 58(4):261–280. https://doi.org/10.1007/s12599-015-0412-2

Kubrak K, Milani F, Nolte A, Dumas M (2022) Prescriptive process monitoring: Quo vadis ? PeerJ Comput Sci 8:e1097. https://doi.org/10.7717/peerj-cs.1097

Kumari M, Kulkarni MS (2022) Developing a prescriptive decision support system for shop floor control. Ind Manage Data Syst 122(8):1853–1881. https://doi.org/10.1108/IMDS-09-2021-0584

Käki A, Kemppainen K, Liesiö J (2019) What to do when decision-makers deviate from model recommendations? Empirical evidence from hydropower industry. Eur J Oper Res 278(3):869–882. https://doi.org/10.1016/j.ejor.2019.04.021

Lash MT, Zhao K (2016) Early predictions of movie success: the who, what, and when of profitability. J Manag Inf Syst 33(3):874–903. https://doi.org/10.1080/07421222.2016.1243969

Lavy S, Garcia JA, Scinto P, Dixit MK (2014) Key performance indicators for facility performance assessment: simulation of core indicators. Constr Manag Econ 32(12):1183–1204. https://doi.org/10.1080/01446193.2014.970208

Leidner DE, Gonzalez E, Koch H (2018) An affordance perspective of enterprise social media and organizational socialization. J Strateg Inf Syst 27(2):117–138. https://doi.org/10.1016/j.jsis.2018.03.003

Lepenioti K, Bousdekis A, Apostolou D, Mentzas G (2020) Prescriptive analytics: literature review and research challenges. Int J Inf Manage 50:57–70. https://doi.org/10.1016/j.ijinfomgt.2019.04.003

Lepenioti K, Bousdekis A, Apostolou D, Mentzas G (2021) Human-augmented prescriptive analytics with interactive multi-objective reinforcement learning. IEEE Access 9:100677–100693. https://doi.org/10.1109/ACCESS.2021.3096662

Levasseur RE (2015) People skills: building analytics decision models that managers use: a change management perspective. Interfaces 45(4):363–364. https://doi.org/10.1287/inte.2015.0798

Leyer M, Oberländer A, Dootson P, Kowalkiewicz M (2020) Patterns of decision-making processes with AI involved Decision-making with artificial intelligence: Towards a novel conceptualization of patterns. In: Pacific Asia Conference on Information Systems (PACIS)

Li X, Zhang L, Xiao T, Zhang S, Chen C (2019) Learning failure modes of soil slopes using monitoring data. Probab Eng Mech 56:50–57. https://doi.org/10.1016/j.probengmech.2019.04.002

Li X, Zhang W, Zhao X, Pu W, Chen P, Liu F (2021) Wartime industrial logistics information integration: framework and application in optimizing deployment and formation of military logistics platforms. J Ind Inf Integr 22:100201. https://doi.org/10.1016/j.jii.2021.100201

Liu B, Lin J, Zhang L, Kumar U (2019) A dynamic prescriptive maintenance model considering system aging and degradation. IEEE Access 7:94931–94943. https://doi.org/10.1109/ACCESS.2019.2928587

Lo VSY, Pachamanova DA (2015) From predictive uplift modeling to prescriptive uplift analytics: a practical approach to treatment optimization while accounting for estimation risk. J Market Anal 3(2):79–95. https://doi.org/10.1057/jma.2015.5

Longo F, Padovano A, De Felice F, Petrillo A, Elbasheer M (2023) From “prepare for the unknown” to “train for what’s coming”: A digital twin-driven and cognitive training approach for the workforce of the future in smart factories. J Ind Inf Integr 32:100437. https://doi.org/10.1016/j.jii.2023.100437

Majchrzak A, Markus ML (2013) Technology Affordances and Constraints in Management Information Systems (MIS). In: Encyclopedia of Management Theory. Sage Publications

Markus ML, Silver M (2008) A foundation for the study of IT effects: a new look at Desanctis and Poole’s concepts of structural features and spirit. J Assoc Inf 9(10):609–632. https://doi.org/10.17705/1jais.00176

Mazon-Olivo B, Hernández-Rojas D, Maza-Salinas J, Pan A (2018) Rules engine and complex event processor in the context of internet of things for precision agriculture. Comput Electron Agric 154:347–360. https://doi.org/10.1016/j.compag.2018.09.013

McFowland E III, Gangarapu S, Bapna R (2021) A prescriptive analytics framework for optimal policy deployment using heterogeneous treatment effects. MIS Q 45(4):1807–1832

Mehdiyev N, Fettke P (2020) PRESCRIPTIVE PROCESS ANALYTICS WITH DEEP LEARNING AND EXPLAINABLE ARTIFICIAL INTELLIGENCE. In: Proceedings of the 28th European Conference on Information Systems (ECIS)

Meske C, Bunde E, Schneider J, Gersch M (2022) Explainable artificial intelligence: objectives, stakeholders, and future research opportunities. Inf Syst Manag 39(1):53–63. https://doi.org/10.1080/10580530.2020.1849465

Mettler T, Sprenger M, Winter R (2017) Service robots in hospitals: new perspectives on niche evolution and technology affordances. Eur J Inf Syst 26(5):451–468. https://doi.org/10.1057/s41303-017-0046-1

Miikkulainen R, Francon O, Meyerson E, Qiu X, Sargent D, Canzani E, Hodjat B (2021) From prediction to prescription: evolutionary optimization of nonpharmaceutical interventions in the COVID-19 pandemic. IEEE Trans Evol Computat 25(2):386–401. https://doi.org/10.1109/TEVC.2021.3063217

Mikalef P, Pappas IO, Krogstie J, Giannakos M (2018) Big data analytics capabilities: a systematic literature review and research agenda. Inf Syst E-Bus Manage 16(3):547–578. https://doi.org/10.1007/s10257-017-0362-y

Mikalef P, Pappas IO, Krogstie J, Pavlou PA (2020) Big data and business analytics: a research agenda for realizing business value. Inf Manag 57(1):103237. https://doi.org/10.1016/j.im.2019.103237

Moore JF, Carvalho A, Davis GA, Abulhassan Y, Megahed FM (2021) Seat assignments with physical distancing in single-destination public transit settings. IEEE Access 9:42985–42993. https://doi.org/10.1109/ACCESS.2021.3065298

Mortenson MJ, Doherty NF, Robinson S (2015) Operational research from taylorism to terabytes: a research agenda for the analytics age. Eur J Oper Res 241(3):583–595. https://doi.org/10.1016/j.ejor.2014.08.029

Nadeem A, Marjanovic O, Abedin B (2022) Gender bias in AI-based decision-making systems: a systematic literature review. AJIS. https://doi.org/10.3127/ajis.v26i0.3835

Nakousi C, Pascual R, Anani A, Kristjanpoller F, Lillo P (2018) An asset-management oriented methodology for mine haul-fleet usage scheduling. Reliab Eng Syst Saf 180:336–344. https://doi.org/10.1016/j.ress.2018.07.034

Niehaus F, Wiesche M (2021) A Socio-Technical Perspective on Organizational Interaction with AI: A Literature Review. In: European Conference on Information Systems (ECIS)

Notz PM (2020) Explainable Subgradient Tree Boosting for Prescriptive Analytics in Operations Management. SSRN Journal https://ssrn.com/abstract=3567665 . https://doi.org/10.2139/ssrn.3567665

Oesterreich TD, Anton E, Teuteberg F, Dwivedi YK (2022) The role of the social and technical factors in creating business value from big data analytics: a meta-analysis. J Bus Res 153:128–149. https://doi.org/10.1016/j.jbusres.2022.08.028

Orlikowski WJ, Iacono CS (2001) Research commentary: desperately seeking the “IT” in IT research: a call to theorizing the IT artifact. Inf Syst Res 12(2):121–134. https://doi.org/10.1287/isre.12.2.121.9700

Pappas IO, Mikalef P, Giannakos MN, Krogstie J, Lekakos G (2018) Big data and business analytics ecosystems: paving the way towards digital transformation and sustainable societies. Inf Syst E-Bus Manage 16(3):479–491. https://doi.org/10.1007/s10257-018-0377-z

Paré G, Trudel M-C, Jaana M, Kitsiou S (2015) Synthesizing information systems knowledge: a typology of literature reviews. Inf Manag 52(2):183–199. https://doi.org/10.1016/j.im.2014.08.008

Peng C, Van Doorn J, Eggers F, Wieringa JE (2022) The effect of required warmth on consumer acceptance of artificial intelligence in service: the moderating role of AI-human collaboration. Int J Inf Manage 66:102533. https://doi.org/10.1016/j.ijinfomgt.2022.102533

Pereira FD, Fonseca SC, Oliveira EHT, Cristea AI, Bellhauser H, Rodrigues L, Oliveira DBF, Isotani S, Carvalho LSG (2021) Explaining individual and collective programming students’ behavior by interpreting a black-box predictive model. IEEE Access 9:117097–117119. https://doi.org/10.1109/ACCESS.2021.3105956

Perugini D, Perugini M (2014) Characterised and personalised predictive-prescriptive analytics using agent-based simulation. IJDATS 6(3):209. https://doi.org/10.1504/IJDATS.2014.063059

Pessach D, Singer G, Avrahami D, Chalutz Ben-Gal H, Shmueli E, Ben-Gal I (2020) Employees recruitment: a prescriptive analytics approach via machine learning and mathematical programming. Decis Support Syst 134:113290. https://doi.org/10.1016/j.dss.2020.113290

Peters M, Ketter W, Saar-Tsechansky M, Collins J (2013) A reinforcement learning approach to autonomous decision-making in smart electricity markets. Mach Learn 92(1):5–39. https://doi.org/10.1007/s10994-013-5340-0

Poornima S, Pushpalatha M (2020) A survey on various applications of prescriptive analytics. Int J Intell Syst 1:76–84. https://doi.org/10.1016/j.ijin.2020.07.001

Pospieszny P (2017) Software estimation: towards prescriptive analytics. In: Proceedings of the 27th International Workshop on Software Measurement and 12th International Conference on Software Process and Product Measurement. ACM, Gothenburg Sweden, pp 221–226

Pozzi G, Pigni F, Vitari C (2014) Affordance Theory in the IS Discipline: a Review and Synthesis of the Literature. In: Americas Conference on Information Systems (AMCIS)

Raeesi Vanani I, Majidian S (2021) Prescriptive Analytics in Internet of Things with Concentration on Deep Learning. In: García Márquez FP, Lev B (eds) Introduction to internet of things in management science and operations research. Springer International Publishing, Cham, pp 31–54

Raychaudhuri SJ, Manjunath S, Srinivasan CP, Swathi N, Sushma S, Nitin Bhushan KN, Narendra Babu C (2021) Prescriptive analytics for impulsive behaviour prevention using real-time biometrics. Prog Artif Intell 10(2):99–112. https://doi.org/10.1007/s13748-020-00229-9

Ren C, Dong J, Ding H, Wang W (2006) Linking Strategic Objectives to Operations: Towards a More Effective Supply Chain Decision Making. In: Proceedings of the 2006 Winter Simulation Conference. pp 1422–1430

Revathy P, Mukesh R (2020) HadoopSec 2.0: Prescriptive analytics-based multi-model sensitivity-aware constraints centric block placement strategy for Hadoop. IFS 39(6):8477–8486

Rider NL, Cahill G, Motazedi T, Wei L, Kurian A, Noroski LM, Seeborg FO, Chinn IK, Roberts K (2021) PI prob: a risk prediction and clinical guidance system for evaluating patients with recurrent infections. PLoS ONE 16(2):e0237285. https://doi.org/10.1371/journal.pone.0237285

Rizzo SG, Chen Y, Pang L, Lucas J, Kaoudi Z, Quiane J, Chawla S (2020) Prescriptive Learning for Air-Cargo Revenue Management. In: IEEE International Conference on Data Mining (ICDM). pp 462–471

Rzepka C, Berger B (2018) User Interaction with AI-enabled Systems: A Systematic Review of IS Research. In: International Conference on Information Systems (ICIS)

Salvio KBV, Palaoag TD (2019) Evaluation of the Selected Philippine E-Government Websites’ Performance with Prescriptive Analysis. In: Proceedings of the 2019 5th International Conference on Computing and Artificial Intelligence - ICCAI ‘19. ACM Press, Bali, Indonesia, pp 129–137

Santos LR, Rosati AG (2015) the evolutionary roots of human decision making. Annu Rev Psychol 66(1):321–347. https://doi.org/10.1146/annurev-psych-010814-015310

Sarker S, Chatterjee S, Xiao X, Elbanna A (2019) The sociotechnical axis of cohesion for the is discipline: its historical legacy and its continued relevance. MIS Q 43(3):695–719. https://doi.org/10.25300/MISQ/2019/13747

Schneider J, Meske C, Kuss P (2024) Foundation models: a new paradigm for artificial intelligence. Bus Inf Syst Eng 66(2):221–231. https://doi.org/10.1007/s12599-024-00851-0

Schoenfeld AH (2010) How we think. Routledge

Book   Google Scholar  

Schoormann T, Strobel G, Möller F, Petrik D, Zschech P (2023) Artificial intelligence for sustainability: a systematic review of information systems literature. CAIS 52:199–237. https://doi.org/10.17705/1CAIS.05209

Sedighi Maman Z, Chen Y-J, Baghdadi A, Lombardo S, Cavuoto LA, Megahed FM (2020) A data analytic framework for physical fatigue management using wearable sensors. Expert Syst Appl 155:113405. https://doi.org/10.1016/j.eswa.2020.113405

Shiau W-L, Chen H, Wang Z, Dwivedi YK (2023) Exploring core knowledge in business intelligence research. Internet Res 33(3):1179–1201. https://doi.org/10.1108/INTR-04-2021-0231

Shollo A, Hopf K, Thiess T, Müller O (2022) Shifting ML value creation mechanisms: a process model of ML value creation. J Strateg Inf Syst 31(3):101734. https://doi.org/10.1016/j.jsis.2022.101734

Silva AJ, Cortez P (2022) An Industry 4.0 Intelligent Decision Support System for Analytical Laboratories. In: Maglogiannis I, Iliadis L, Macintyre J, Cortez P (eds) Artificial intelligence applications and innovations. Springer International Publishing, Cham, pp 159–169

Simon HA (1960) The new science of management decision. Harper & Brothers, New York

Slovic P, Fischhoff B, Lichtenstein S (1977) Behavioral decision theory. Annu Rev Psychol 28(1):1–39. https://doi.org/10.1146/annurev.ps.28.020177.000245

Soeffker N, Ulmer MW, Mattfeld DC (2022) Stochastic dynamic vehicle routing in the light of prescriptive analytics: a review. Eur J Oper Res 298(3):801–820. https://doi.org/10.1016/j.ejor.2021.07.014

Song S, Jeong D-H, Kim J, Hwang M, Gim J, Jung H (2014) Research Advising System Based on Prescriptive Analytics. In: Park JJ, Pan Y, Kim C-S, Yang Y (eds) Future information technology. Springer, Berlin Heidelberg, Berlin, Heidelberg, pp 569–574

Soroush M, Masooleh LS, Seider WD, Oktem U, Arbogast JE (2020) Model-predictive safety optimal actions to detect and handle process operation hazards. AIChE J. https://doi.org/10.1002/aic.16932

Srinivas S, Ravindran AR (2018) Optimizing outpatient appointment system using machine learning algorithms and scheduling rules: a prescriptive analytics framework. Expert Syst Appl 102:245–261. https://doi.org/10.1016/j.eswa.2018.02.022

Stefani K, Zschech P (2018) Constituent Elements for Prescriptive Analytics Systems. In: Twenty-Sixth European Conference on Information Systems

Stein N, Meller J, Flath CM (2018) Big data on the shop-floor: sensor-based decision-support for manual processes. J Bus Econ 88(5):593–616. https://doi.org/10.1007/s11573-017-0890-4

Straub D, Welpe I (2014) Decision-Making Under Risk: A Normative and Behavioral Perspective. In: Klüppelberg C, Straub D, Welpe IM (eds) risk: a multidisciplinary introduction. Springer International Publishing, Cham, pp 63–93

Strong D, Volkoff O, Johnson S, Pelletier L, Tulu B, Bar-On I, Trudel J, Garber L (2014) A theory of organization-EHR affordance actualization. J Assoc Inf 15(2):53–85. https://doi.org/10.17705/1jais.00353

Sturm T, Gerlacha J, Pumplun L, Mesbah N, Peters F, Tauchert C, Nan N, Buxmann P (2021) Coordinating human and machine learning for effective organization learning. MIS Q 45(3):1581–1602. https://doi.org/10.25300/MISQ/2021/16543

Suvarna M, Jahirul MI, Aaron-Yeap WH, Augustine CV, Umesh A, Rasul MG, Günay ME, Yildirim R, Janaun J (2022) Predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning. Renew Energ 189:245–258. https://doi.org/10.1016/j.renene.2022.02.124

Svenson O (1992) Differentiation and consolidation theory of human decision making: a frame of reference for the study of pre-and post-decision processes. Acta Psychol 80(1–3):143–168. https://doi.org/10.1016/0001-6918(92)90044-E

Swaminathan JM (2018) Big data analytics for rapid, impactful, sustained, and efficient (RISE) humanitarian operations. Prod Oper Manag 27(9):1696–1700. https://doi.org/10.1111/poms.12840

Tamimi N, Samani S, Minaei M, Harirchi F (2019) An Artificial Intelligence Decision Support System for Unconventional Field Development Design. In: Proceedings of the 7th Unconventional Resources Technology Conference

Tektaş B, Turan HH, Kasap N, Çebi F, Delen D (2022) A fuzzy prescriptive analytics approach to power generation capacity planning. Energies 15(9):3176. https://doi.org/10.3390/en15093176

Thapa D, Sein MK (2018) Trajectory of affordances: insights from a case of telemedicine in Nepal. Info Systems J 28(5):796–817. https://doi.org/10.1111/isj.12160

Trunk A, Birkel H, Hartmann E (2020) On the current state of combining human and artificial intelligence for strategic organizational decision making. Bus Res 13(3):875–919. https://doi.org/10.1007/s40685-020-00133-x

Uskov VL, Bakken JP, Shah A, Hancher N, McPartlin C, Gayke K (2019) Innovative InterLabs System for Smart Learning Analytics in Engineering Education. 2019 IEEE global engineering education conference (EDUCON). IEEE, Dubai, United Arab Emirates, pp 1363–1369

Vanderschueren T, Verdonck T, Baesens B, Verbeke W (2022) Predict-then-optimize or predict-and-optimize? an empirical evaluation of cost-sensitive learning strategies. Inf Sci 594:400–415. https://doi.org/10.1016/j.ins.2022.02.021

Vargas AP, Díaz D, Jaramillo S, Rangel F, Villa D, Villegas JG (2022) Improving the tactical planning of solid waste collection with prescriptive analytics: a case study. Prod 32:e20210037. https://doi.org/10.1590/0103-6513.20210037

Vater J, Harscheidt L, Knoll A (2019) A Reference Architecture Based on Edge and Cloud Computing for Smart Manufacturing. In: 28th International Conference on Computer Communication and Networks (ICCCN)

Vater J, Schlaak P, Knoll A (2020) A Modular Edge-/Cloud-Solution for Automated Error Detection of Industrial Hairpin Weldings using Convolutional Neural Networks. In: IEEE 44th Annual Computers, Software, and Applications Conference (COMPSAC). pp 505–510

Verbraeken J, Wolting M, Katzy J, Kloppenburg J, Verbelen T, Rellermeyer JS (2021) A survey on distributed machine learning. ACM Comput Surv 53(2):1–33. https://doi.org/10.1145/3377454

Vieira AAC, Dias LMS, Santos MY, Pereira GAB, Oliveira JA (2020) Supply chain data integration: a literature review. J Ind Inf Integr 19:100161. https://doi.org/10.1016/j.jii.2020.100161

Vimala Rani M, Mathirajan M (2021) Prescriptive Analytics for Dynamic Real Time Scheduling of Diffusion Furnaces. In: Srinivas S, Rajendran S, Ziegler H (eds) Supply Chain Management in manufacturing and service systems. Springer International Publishing, Cham, pp 241–278

Wang C-H, Cheng H-Y, Deng Y-T (2018) Using bayesian belief network and time-series model to conduct prescriptive and predictive analytics for computer industries. Comput Ind Eng 115:486–494. https://doi.org/10.1016/j.cie.2017.12.003

Wang N, Ho K, Pavlou G, Howarth M (2008) An overview of routing optimization for internet traffic engineering. IEEE Commun Surv Tutorials 10(1):36–56. https://doi.org/10.1109/COMST.2008.4483669

Wanner J, Herm L-V, Heinrich K, Janiesch C (2022) The effect of transparency and trust on intelligent system acceptance: evidence from a user-based study. Electron Mark 32(4):2079–2102. https://doi.org/10.1007/s12525-022-00593-5

Wanner J, Wissuchek C, Welsch G, Janiesch C (2023) A taxonomy and archetypes of business analytics in smart manufacturing. Data Base Adv Inf Syst 54(1):11–45. https://doi.org/10.1145/3583581.3583584

Webster J, Watson RT (2002) Analyzing the Past to Prepare for the Future: Writing a Literature Review. MIS Q 26(2):xiii–xxiii

Xylia M, Ibrahim O, Silveira S (2016) Fossil-free public transport: Prescriptive policy analysis for the Swedish bus fleets. 2016 13th International Conference on the European Energy Market (EEM). IEEE, Porto, Portugal, pp 1–5

Google Scholar  

Yang H, Duque D, Morton DP (2022) Optimizing diesel fuel supply chain operations to mitigate power outages for hurricane relief. IISE Transactions 54(10):936–949. https://doi.org/10.1080/24725854.2021.2021461

Yanta S, Thammaboosadee S, Chanyagorn P, Chuckpaiwong R (2021) Probation Status Prediction and Optimization for Undergraduate Engineering Students. 2021 13th International Conference on Knowledge and Smart Technology (KST). IEEE, Bangsaen, Chonburi, Thailand, pp 191–196

Zadorojniy A, Wasserkrug S, Zeltyn S, Lipets V (2019) Unleashing analytics to reduce costs and improve quality in wastewater treatment. INFORMS J on Appl Anal 49(4):262–268. https://doi.org/10.1287/inte.2019.0990

Zhang P, Zhu X, Xie M (2021) A model-based reinforcement learning approach for maintenance optimization of degrading systems in a large state space. Comput Ind Eng 161:107622. https://doi.org/10.1016/j.cie.2021.107622

Zhang B, Ong YJ, Nakamura T (2022) SimPO: Simultaneous Prediction and Optimization. In: 2022 IEEE International Conference on Services Computing (SCC). IEEE, pp 120–122

Zheng H, Ryzhov IO, Xie W, Zhong J (2021) Personalized multimorbidity management for patients with type 2 diabetes using reinforcement learning of electronic health records. Drugs 81(4):471–482. https://doi.org/10.1007/s40265-020-01435-4

Zschech P, Weinzierl S, Hambauer N, Zilker S, Kraus M (2022) GAM(e) changer or not? An evaluation of interpretable machine learning models based on additive model constraints. In: Proceedings of the 30th European Conference on Information Systems (ECIS)

vom Brocke J, Simons A, Niehaves B, Riemer K, Plattfat R, Cleven A (2009) RECONSTRUCTING THE GIANT: ON THE IMPORTANCE OF RIGOUR IN DOCUMENTING THE LITERATURE SEARCH PROCESS. In: European Conference on Information Systems (ECIS)

Download references

Open Access funding enabled and organized by Projekt DEAL.

Author information

Authors and affiliations.

FAU Erlangen-Nürnberg, Lange Gasse 20, 90403, Nuremberg, Germany

Christopher Wissuchek

Universität Leipzig, Grimmaische Str. 12, 04109, Leipzig, Germany

Patrick Zschech

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Christopher Wissuchek .

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix A: Detailed search syntax

Appendix B: Literature sample

Appendix C: Identified constituent components in literature sample

Appendix D: Identified technology affordances in literature sample

Appendix E: Identified system archetypes

Appendix F: Decision processing techniques

In the following, we give an overview of exemplary technique subcategories used in prescriptive analytics, following the survey of Lepenioti et al. ( 2020 ).

Appendix G: Overview of data properties

Appendix H: Temporal trend analysis of results

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Wissuchek, C., Zschech, P. Prescriptive analytics systems revised: a systematic literature review from an information systems perspective. Inf Syst E-Bus Manage (2024). https://doi.org/10.1007/s10257-024-00688-w

Download citation

Received : 15 October 2023

Revised : 31 May 2024

Accepted : 09 August 2024

Published : 30 August 2024

DOI : https://doi.org/10.1007/s10257-024-00688-w

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Prescriptive analytics
• Business analytics
• Decision support system
• Systematic literature review
• Decision-making
• Artificial intelligence
• Find a journal
• Publish with us
• Track your research

Information

• Author Services

Initiatives

You are accessing a machine-readable page. In order to be human-readable, please install an RSS reader.

All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. For more information, please refer to https://www.mdpi.com/openaccess .

Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications.

Feature papers are submitted upon individual invitation or recommendation by the scientific editors and must receive positive feedback from the reviewers.

Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to readers, or important in the respective research area. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.

Original Submission Date Received: .

• Active Journals
• Find a Journal
• Proceedings Series
• For Authors
• For Reviewers
• For Editors
• For Librarians
• For Publishers
• For Societies
• For Conference Organizers
• Open Access Policy
• Institutional Open Access Program
• Special Issues Guidelines
• Editorial Process
• Research and Publication Ethics
• Article Processing Charges
• Testimonials
• Preprints.org
• SciProfiles
• Encyclopedia

Article Menu

• Subscribe SciFeed
• Recommended Articles
• Google Scholar
• on Google Scholar
• Table of Contents

Find support for a specific problem in the support section of our website.

Please let us know what you think of our products and services.

Visit our dedicated information section to learn more about MDPI.

JSmol Viewer

Oxygen vacancy engineering and its impact on resistive switching of oxide thin films for memory and neuromorphic applications.

1. Introduction

2. rs switching mode and parameters, 3. resistive switching mechanism, 4. modulation of the concentration of v o s in metal oxides, 4.1. doping, 4.2. magnetron sputtering technique, 4.3. thermal treatment, 4.4. pulsed laser deposition (pld), 4.5. atomic layer deposition (ald), 5. oxygen vacancy-engineered oxide thin films for memristors, 5.1. memory storage applications, 5.2. neuromorphic applications, 6. summary and future perspectives, author contributions, conflicts of interest.

• Chua, L. Memristor-the missing circuit element. IEEE Trans. Circuit Theory 1971 , 18 , 507–519. [ Google Scholar ] [ CrossRef ]
• Chua, L. Everything you wish to know about memristors but are afraid to ask. In Handbook of Memristor Networks ; Springer: Cham, Switzerland, 2019; pp. 89–157. [ Google Scholar ]
• Xiao, Y.; Jiang, B.; Zhang, Z.; Ke, S.; Jin, Y.; Wen, X. A review of memristor: Material and structure design, device performance, applications and prospects. Sci. Technol. Adv. Mater. 2023 , 24 , 2162323. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Hickmott, T.W. Low-frequency negative resistance in thin anodic oxide films. J. Appl. Phys. 1962 , 33 , 2669–2682. [ Google Scholar ] [ CrossRef ]
• Liu, S.; Wu, N.J.; Ignatiev, A. Electric-pulse-induced reversible resistance change effect in magnetoresistive films. Appl. Phys. Lett. 2000 , 76 , 2749–2751. [ Google Scholar ] [ CrossRef ]
• Lee, M.-J.; Park, Y.; Kang, B.-S.; Ahn, S.-E.; Lee, C.; Kim, K.; Xianyu, W.; Stefanovich, G.; Lee, J.-H.; Chung, S.-J. 2-stack 1D-1R cross-point structure with oxide diodes as switch elements for high density resistance RAM applications. In Proceedings of the 2007 IEEE International Electron Devices Meeting, Washington, DC, USA, 10–12 December 2007; pp. 771–774. [ Google Scholar ]
• Strukov, D.B.; Snider, G.S.; Stewart, D.R.; Williams, R.S. The missing memristor found. Nature 2008 , 453 , 80–83. [ Google Scholar ] [ CrossRef ]
• Chevallier, C.J.; Siau, C.H.; Lim, S.F.; Namala, S.R.; Matsuoka, M.; Bateman, B.L.; Rinerson, D. A 0.13 µm 64 Mb multi-layered conductive metal-oxide memory. In Proceedings of the 2010 IEEE International Solid-State Circuits Conference, San Francisco, CA, USA, 7–11 February 2010; pp. 260–261. [ Google Scholar ]
• Liu, T.-Y.; Yan, T.H.; Scheuerlein, R.; Chen, Y.; Lee, J.K.; Balakrishnan, G.; Yee, G.; Zhang, H.; Yap, A.; Ouyang, J. A 130.7 mm 2 2-layer 32 Gb ReRAM memory device in 24 nm technology. In Proceedings of the 2013 IEEE International Solid-State Circuits Conference Digest of Technical Papers, San Francisco, CA, USA, 17–21 February 2013. [ Google Scholar ]
• Fackenthal, R. A 16 GB ReRAM with 200 MB/s write and 1 GB/s read in 27 nm Technology. In Proceedings of the 2014 IEEE International Solid-State Circuits Conference Digest of Technical Papers (ISSCC), San Francisco, CA, USA, 9–13 February 2014. [ Google Scholar ]
• Luo, Q.; Xu, X.; Liu, H.; Lv, H.; Gong, T.; Long, S.; Liu, Q.; Sun, H.; Banerjee, W.; Li, L. Super non-linear RRAM with ultra-low power for 3D vertical nano-crossbar arrays. Nanoscale 2016 , 8 , 15629–15636. [ Google Scholar ] [ CrossRef ]
• Clarke, P. TSMC Offers 22 nm RRAM, Taking MRAM on to 16 nm. Available online: https://www.eenewseurope.com/en/tsmc-offers-22nm-rram-taking-mram-on-to-16nm/ (accessed on 25 August 2020).
• Nano, W. Weebit Nano Taped-Out Its First 22 nm Demo RRAM Chip. Available online: https://www.rram-info.com/weebit-nano-taped-out-its-first-22-nm-demo-rram-chip (accessed on 5 January 2023).
• Wang, C.; Shi, G.; Qiao, F.; Lin, R.; Wu, S.; Hu, Z. Research progress in architecture and application of RRAM with computing-in-memory. Nanoscale Adv. 2023 , 5 , 1559–1573. [ Google Scholar ] [ CrossRef ]
• Gao, B.; Wu, H.; Wu, W.; Wang, X.; Yao, P.; Xi, Y.; Zhang, W.; Deng, N.; Huang, P.; Liu, X. Modeling disorder effect of the oxygen vacancy distribution in filamentary analog RRAM for neuromorphic computing. In Proceedings of the 2017 IEEE International Electron Devices Meeting (IEDM), San Francisco, CA, USA, 2–6 December 2017; p. 4. [ Google Scholar ]
• Banerjee, W.; Liu, Q.; Hwang, H. Engineering of defects in resistive random access memory devices. J. Appl. Phys. 2020 , 127 , 051101. [ Google Scholar ] [ CrossRef ]
• Prakash, A.; Jana, D.; Maikap, S. TaO x -based resistive switching memories: Prospective and challenges. Nanoscale Res. Lett. 2013 , 8 , 418. [ Google Scholar ] [ CrossRef ]
• Asif, M.; Kumar, A. Resistive switching in emerging materials and their characteristics for neuromorphic computing. Mater. Today Electron. 2022 , 1 , 100004. [ Google Scholar ] [ CrossRef ]
• Lashkare, S.; Uddin, W.; Priyadarshi, K.; Ganguly, U. Emerging Memory Technologies for Data Storage and Brain—Inspired Computation: A Global View with Indian Research Insights with a Focus on Resistive Memories. Proc. Natl. Acad. Sci. India Sect. A Phys. Sci. 2023 , 93 , 459–476. [ Google Scholar ] [ CrossRef ]
• Kozicki, M.N.; Mitkova, M.; Valov, I. Electrochemical metallization memories. In Resistive Switching: From Fundamentals of Nanoionic Redox Processes to Memristive Device Applications ; Wiley: Weinheim, Germany, 2016; pp. 483–514. [ Google Scholar ]
• Ielmini, D.; Bruchhaus, R.; Waser, R. Thermochemical resistive switching: Materials, mechanisms, and scaling projections. Phase Transit. 2011 , 84 , 570–602. [ Google Scholar ] [ CrossRef ]
• Zhao, L.; Ryu, S.-W.; Hazeghi, A.; Duncan, D.; Magyari-Köpe, B.; Nishi, Y. Dopant selection rules for extrinsic tunability of HfO x RRAM characteristics: A systematic study. In Proceedings of the 2013 Symposium on VLSI Technology, Kyoto, Japan, 11–13 June 2013; pp. T106–T107. [ Google Scholar ]
• Yang, Y.; Lu, W.D. Progress in the characterizations and understanding of conducting filaments in resistive switching devices. IEEE Trans. Nanotechnol. 2016 , 15 , 465–472. [ Google Scholar ] [ CrossRef ]
• Jiang, H.; Stewart, D.A. Enhanced oxygen vacancy diffusion in Ta 2 O 5 resistive memory devices due to infinitely adaptive crystal structure. J. Appl. Phys. 2016 , 119 , 134502. [ Google Scholar ] [ CrossRef ]
• Lee, J.; Schell, W.; Zhu, X.; Kioupakis, E.; Lu, W.D. Charge Transition of Oxygen Vacancies during Resistive Switching in Oxide-Based RRAM. ACS Appl. Mater. Interfaces 2019 , 11 , 11579–11586. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Jiang, H.; Stewart, D.A. Using dopants to tune oxygen vacancy formation in transition metal oxide resistive memory. ACS Appl. Mater. Interfaces 2017 , 9 , 16296–16304. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Kukli, K.; Kemell, M.; Vehkamäki, M.; Heikkilä, M.J.; Mizohata, K.; Kalam, K.; Ritala, M.; Leskelä, M.; Kundrata, I.; Fröhlich, K. Atomic layer deposition and properties of mixed Ta 2 O 5 and ZrO 2 films. AIP Adv. 2017 , 7 , 025001. [ Google Scholar ] [ CrossRef ]
• Zhu, X.; Zhuge, F.; Li, M.; Yin, K.; Liu, Y.; Zuo, Z.; Chen, B.; Li, R.-W. Microstructure dependence of leakage and resistive switching behaviours in Ce-doped BiFeO 3 thin films. J. Phys. D Appl. Phys. 2011 , 44 , 415104. [ Google Scholar ] [ CrossRef ]
• Lei, M.; He, H.; Yu, Q.; Chen, C.; Lu, Y.; Ye, Z. Optical properties of Na-doped ZnO nanorods grown by metalorganic chemical vapor deposition. Mater. Lett. 2015 , 160 , 547–549. [ Google Scholar ] [ CrossRef ]
• Chang, T.-J.; Li, C.-Y.; Chu, S.-Y. Ta 2 O 5 doping effects on the property improvement of HfOx-based RRAMs using co-sputtering deposition method. Mater. Charact. 2023 , 199 , 112786. [ Google Scholar ] [ CrossRef ]
• Zhang, P.; Gao, C.; Lv, F.; Wei, Y.; Dong, C.; Jia, C.; Liu, Q.; Xue, D. Hydrothermal epitaxial growth and nonvolatile bipolar resistive switching behavior of LaFeO 3 -PbTiO 3 films on Nb: SrTiO 3 (001) substrate. Appl. Phys. Lett. 2014 , 105 , 15. [ Google Scholar ]
• Henning, R.A.; Leichtweiss, T.; Dorow-Gerspach, D.; Schmidt, R.; Wolff, N.; Schürmann, U.; Decker, Y.; Kienle, L.; Wuttig, M.; Janek, J. Phase formation and stability in TiO x and ZrO x thin films: Extremely sub-stoichiometric functional oxides for electrical and TCO applications. Z. Krist. Mater. 2017 , 232 , 161–183. [ Google Scholar ]
• Mundle, R.; Carvajal, C.; Pradhan, A.K. ZnO/Al: ZnO transparent resistive switching devices grown by atomic layer deposition for memristor applications. Langmuir 2016 , 32 , 4983–4995. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Deepa, S.; Kumari, K.P.; Thomas, B. Contribution of oxygen-vacancy defect-types in enhanced CO 2 sensing of nanoparticulate Zn-doped SnO 2 films. Ceram. Int. 2017 , 43 , 17128–17141. [ Google Scholar ] [ CrossRef ]
• Cai, Q.; Duan, Z.; Chen, J.; Wang, X.; Zhu, W.; Liu, S.; Xiao, P.; Yu, X. Tristable TaOx-based memristor by controlling oxygen vacancy transportion based on valence transition mechanism. Ceram. Int. 2024 , in press . [ Google Scholar ] [ CrossRef ]
• Pan, X.; Yang, M.-Q.; Fu, X.; Zhang, N.; Xu, Y.-J. Defective TiO 2 with oxygen vacancies: Synthesis, properties and photocatalytic applications. Nanoscale 2013 , 5 , 3601–3614. [ Google Scholar ] [ CrossRef ]
• Abbas, Y.; Han, I.S.; Sokolov, A.S.; Jeon, Y.R.; Choi, C. Rapid thermal annealing on the atomic layer-deposited zirconia thin film to enhance resistive switching characteristics. J. Mater. Sci. Mater. Electron. 2020 , 31 , 903–909. [ Google Scholar ] [ CrossRef ]
• Liu, C.; Zhang, C.-C.; Cao, Y.-Q.; Wu, D.; Wang, P.; Li, A.-D. Optimization of oxygen vacancy concentration in HfO 2 /HfO x bilayer-structured ultrathin memristors by atomic layer deposition and their biological synaptic behavior. J. Mater. Chem. C 2020 , 8 , 12478–12484. [ Google Scholar ] [ CrossRef ]
• Sharath, S.U.; Bertaud, T.; Kurian, J.; Hildebrandt, E.; Walczyk, C.; Calka, P.; Zaumseil, P.; Sowinska, M.; Walczyk, D.; Gloskovskii, A.; et al. Towards forming-free resistive switching in oxygen engineered HfO 2−x . Appl. Phys. Lett. 2014 , 104 , 063502. [ Google Scholar ] [ CrossRef ]
• Stevens, J.E.; Lohn, A.J.; Decker, S.A.; Doyle, B.L.; Mickel, P.R.; Marinella, M.J. Reactive sputtering of substoichiometric Ta 2 O x for resistive memory applications. J. Vac. Sci. Technol. A Vac. Surf. Film. 2014 , 32 , 021501. [ Google Scholar ] [ CrossRef ]
• Sharath, S.U.; Joseph, M.J.; Vogel, S.; Hildebrandt, E.; Komissinskiy, P.; Kurian, J.; Schröder, T.; Alff, L. Impact of oxygen stoichiometry on electroforming and multiple switching modes in TiN/TaO x /Pt based ReRAM. Appl. Phys. Lett. 2016 , 109 , 173503. [ Google Scholar ] [ CrossRef ]
• Yang, J.J.; Strukov, D.B.; Stewart, D.R. Memristive devices for computing. Nat. Nanotechnol. 2013 , 8 , 13–24. [ Google Scholar ] [ CrossRef ]
• Landon, C.D.; Wilke, R.H.T.; Brumbach, M.T.; Brennecka, G.L.; Blea-Kirby, M.; Ihlefeld, J.F.; Marinella, M.J.; Beechem, T.E. Thermal transport in tantalum oxide films for memristive applications. Appl. Phys. Lett. 2015 , 107 , 023108. [ Google Scholar ] [ CrossRef ]
• Egorov, K.V.; Kuzmichev, D.S.; Chizhov, P.S.; Lebedinskii, Y.Y.; Hwang, C.S.; Markeev, A.M. In situ control of oxygen vacancies in TaO x thin films via plasma-enhanced atomic layer deposition for resistive switching memory applications. ACS Appl. Mater. Interfaces 2017 , 9 , 13286–13292. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Prakash, A.; Maikap, S.; Chiu, H.C.; Tien, T.C.; Lai, C.S. Enhanced resistive switching memory characteristics and mechanism using a Ti nanolayer at the W/TaO x interface. Nanoscale Res. Lett. 2013 , 8 , 125. [ Google Scholar ]
• Rudrapal, K.; Bhattacharya, G.; Adyam, V.; Chaudhuri, A.R. Forming-free, self-compliance, bipolar multi-level resistive switching in WO3–x based MIM device. Adv. Electron. Mater. 2022 , 8 , 2200250. [ Google Scholar ] [ CrossRef ]
• Rudrapal, K.; Biswas, M.; Jana, B.; Adyam, V.; Chaudhuri, A.R. Tuning resistive switching properties of WO3− x-memristors by oxygen vacancy engineering for neuromorphic and memory storage applications. J. Phys. D Appl. Phys. 2023 , 56 , 205302. [ Google Scholar ] [ CrossRef ]
• Ghenzi, N.; Rozenberg, M.J.; Llopis, R.; Levy, P.; Hueso, L.E.; Stoliar, P. Tuning the resistive switching properties of TiO 2− x films. Appl. Phys. Lett. 2015 , 106 , 123509. [ Google Scholar ] [ CrossRef ]
• Li, C.-Y.; Lin, C.-C.; Chu, S.-Y.; Lin, J.-T.; Huang, C.-Y.; Hong, C.-S. Effects of Nb doping on switching-voltage stability of zinc oxide thin films. J. Appl. Phys. 2020 , 128 , 175308. [ Google Scholar ] [ CrossRef ]
• Li, H.; Chen, Q.; Chen, X.; Mao, Q.; Xi, J.; Ji, Z. Improvement of resistive switching in ZnO film by Ti doping. Thin Solid Film. 2013 , 537 , 279–284. [ Google Scholar ] [ CrossRef ]
• Xu, D.; Xiong, Y.; Tang, M.; Zeng, B. Coexistence of the bipolar and unipolar resistive switching behaviors in vanadium doped ZnO films. J. Alloys Compd. 2014 , 584 , 269–272. [ Google Scholar ] [ CrossRef ]
• Xu, D.L.; Xiong, Y.; Tang, M.H.; Zeng, B.W.; Xiao, Y.G. Bipolar and unipolar resistive switching modes in Pt/Zn0.99Zr0.01O/Pt structure for multi-bit resistance random access memory. Appl. Phys. Lett. 2014 , 104 , 183501. [ Google Scholar ] [ CrossRef ]
• Sedghi, N.; Li, H.; Brunell, I.F.; Dawson, K.; Potter, R.J.; Guo, Y.; Gibbon, J.T.; Dhanak, V.R.; Zhang, W.D.; Zhang, J.F. The role of nitrogen doping in ALD Ta 2 O 5 and its influence on multilevel cell switching in RRAM. Appl. Phys. Lett. 2017 , 110 , 102902. [ Google Scholar ] [ CrossRef ]
• Rasool, A.; Amiruddin, R.; Mohamed, I.R.; Kumar, M.C.S. Fabrication and characterization of resistive random access memory (ReRAM) devices using molybdenum trioxide (MoO 3 ) as switching layer. Superlattices Microstruct. 2020 , 147 , 106682. [ Google Scholar ] [ CrossRef ]
• Xu, J.; Wang, H.; Zhu, Y.; Liu, Y.; Zou, Z.; Li, G.; Xiong, R. Tunable digital-to-analog switching in Nb 2 O 5 -based resistance switching devices by oxygen vacancy engineering. Appl. Surf. Sci. 2022 , 579 , 152114. [ Google Scholar ] [ CrossRef ]
• Swathi, S.P.; Angappane, S. Enhanced resistive switching performance of hafnium oxide-based devices: Effects of growth and annealing temperatures. J. Alloys Compd. 2022 , 913 , 165251. [ Google Scholar ] [ CrossRef ]
• Loy, D.J.J.; Dananjaya, P.A.; Chakrabarti, S.; Tan, K.H.; Chow, S.C.W.; Toh, E.H.; Lew, W.S. Oxygen vacancy density dependence with a hopping conduction mechanism in multilevel switching behavior of HfO 2 -based resistive random access memory devices. ACS Appl. Electron. Mater. 2020 , 2 , 3160–3170. [ Google Scholar ] [ CrossRef ]
• Zhang, R.; Huang, H.; Xia, Q.; Ye, C.; Wei, X.; Wang, J.; Zhang, L.; Zhu, L.Q. Role of oxygen vacancies at the TiO 2 /HfO 2 interface in flexible oxide-based resistive switching memory. Adv. Electron. Mater. 2019 , 5 , 1800833. [ Google Scholar ] [ CrossRef ]
• Banerjee, W.; Kashir, A.; Kamba, S. Hafnium oxide (HfO 2 )—A multifunctional oxide: A review on the prospect and challenges of hafnium oxide in resistive switching and ferroelectric memories. Small 2022 , 18 , 2107575. [ Google Scholar ] [ CrossRef ]
• Goux, L.; Fantini, A.; Kar, G.; Chen, Y.-Y.; Jossart, N.; Degraeve, R.; Clima, S.; Govoreanu, B.; Lorenzo, G.; Pourtois, G. Ultralow sub-500nA operating current high-performance TiN\Al 2 O 3 \HfO 2 \Hf\TiN bipolar RRAM achieved through understanding-based stack-engineering. In Proceedings of the 2012 Symposium on VLSI Technology (VLSIT), Honolulu, HI, USA, 12–14 June 2012; pp. 159–160. [ Google Scholar ]
• Wan, C.J.; Liu, Y.H.; Zhu, L.Q.; Feng, P.; Shi, Y.; Wan, Q. Short-term synaptic plasticity regulation in solution-gated indium–gallium–zinc-oxide electric-double-layer transistors. ACS Appl. Mater. Interfaces 2016 , 8 , 9762–9768. [ Google Scholar ] [ CrossRef ]
• He, Y.; Yang, Y.; Nie, S.; Liu, R.; Wan, Q. Electric-double-layer transistors for synaptic devices and neuromorphic systems. J. Mater. Chem. C 2018 , 6 , 5336–5352. [ Google Scholar ] [ CrossRef ]
• Abbas, Y.; Jeon, Y.-R.; Sokolov, A.S.; Kim, S.; Ku, B.; Choi, C. Compliance-free, digital SET and analog RESET synaptic characteristics of sub-tantalum oxide based neuromorphic device. Sci. Rep. 2018 , 8 , 1228. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Wang, Z.; Yin, M.; Zhang, T.; Cai, Y.; Wang, Y.; Yang, Y.; Huang, R. Engineering incremental resistive switching in TaO x based memristors for brain-inspired computing. Nanoscale 2016 , 8 , 14015–14022. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Gao, B.; Liu, L.; Kang, J. Investigation of the synaptic device based on the resistive switching behavior in hafnium oxide. Prog. Nat. Sci. 2015 , 25 , 47–50. [ Google Scholar ] [ CrossRef ]
• Gao, B.; Kang, J.F.; Chen, Y.S.; Zhang, F.F.; Chen, B.; Huang, P.; Liu, L.F.; Liu, X.Y.; Wang, Y.Y.; Tran, X.A. Oxide-based RRAM: Unified microscopic principle for both unipolar and bipolar switching. In Proceedings of the 2011 International Electron Devices Meeting, Washington, DC, USA, 5–7 December 2011; pp. 14–17. [ Google Scholar ]
• Mohapatra, A.; Mhaskar, C.M.; Sahu, M.C.; Sahoo, S.; Chaudhuri, A.R. Neuromorphic Learning and Recognition in WO3-x Thin Film-based Forming-free Flexible Electronic Synapses. Nanotechnology 2024 , 35 , 455702. [ Google Scholar ] [ CrossRef ]
• Qin, F.; Zhang, Y.; Song, H.W.; Lee, S. Enhancing memristor fundamentals through instrumental characterization and understanding reliability issues. Mater. Adv. 2023 , 4 , 1850–1875. [ Google Scholar ] [ CrossRef ]
• Pérez, E.; Maldonado, D.; Acal, C.; Ruiz-Castro, J.E.; Alonso, F.J.; Aguilera, A.M.; Jiménez-Molinos, F.; Wenger, C.; Roldán, J.B. Analysis of the statistics of device-to-device and cycle-to-cycle variability in TiN/Ti/Al: HfO 2 /TiN RRAMs. Microelectron. Eng. 2019 , 214 , 104–109. [ Google Scholar ] [ CrossRef ]
• Akinaga, H.; Shima, H. Resistive random access memory (ReRAM) based on metal oxides. Proc. IEEE 2010 , 98 , 2237–2251. [ Google Scholar ] [ CrossRef ]
• Fadeev, A.V.; Rudenko, K.V. To the issue of the memristor’s hrs and lrs states degradation and data retention time. Russ. Microelectron 2021 , 50 , 311–325. [ Google Scholar ] [ CrossRef ]
• Ravi, V. Review of memristor based neuromorphic computation: Opportunities, challenges and applications. Eng. Res. Express 2024 , 6 , 032203. [ Google Scholar ]
• Wang, S.; Song, L.; Chen, W.; Wang, G.; Hao, E.; Li, C.; Hu, Y.; Pan, Y.; Nathan, A.; Hu, G. Memristor-Based Intelligent Human-Like Neural Computing. Adv. Electron. Mater. 2023 , 9 , 2200877. [ Google Scholar ] [ CrossRef ]

Click here to enlarge figure

Share and Cite

Jana, B.; Roy Chaudhuri, A. Oxygen Vacancy Engineering and Its Impact on Resistive Switching of Oxide Thin Films for Memory and Neuromorphic Applications. Chips 2024 , 3 , 235-257. https://doi.org/10.3390/chips3030012

Jana B, Roy Chaudhuri A. Oxygen Vacancy Engineering and Its Impact on Resistive Switching of Oxide Thin Films for Memory and Neuromorphic Applications. Chips . 2024; 3(3):235-257. https://doi.org/10.3390/chips3030012

Jana, Biswajit, and Ayan Roy Chaudhuri. 2024. "Oxygen Vacancy Engineering and Its Impact on Resistive Switching of Oxide Thin Films for Memory and Neuromorphic Applications" Chips 3, no. 3: 235-257. https://doi.org/10.3390/chips3030012

Article Metrics

Article access statistics, further information, mdpi initiatives, follow mdpi.

Subscribe to receive issue release notifications and newsletters from MDPI journals