Multi-unit organizations are a form of organizations where the geographically dispersed units provide similar products or services in different markets. Deciding on an appropriate level of centralization in such organizations presents a unique challenge. One the one hand the organizations want to maintain a consistent brand identity in all units through centralized control, but on the other hand, they want to provide the units with sufficient autonomy to respond to the challenges they face locally. Traditionally, this challenge was perceived to require a trade-off between performance and standardization, with performance demanding more decentralization and standardization requiring more centralized control. However, our research explores how organizations can potentially resolve this trade-off by promoting norms for knowledge-sharing and setting up the right communication channels, relying on the unit managers’ intrinsic tendency to conform to the behavior of their peers. We build an agent-based model of an organization with multiple interdependent units facing highly similar task environments to investigate how unit managers’ ability to communicate, share knowledge, and conform to peer practices might influence organizational dynamics. We find that, under specific communication network structures, increased decentralization can enhance both performance and standardization without sacrificing one or the other. Furthermore, we discover that centralization might still be preferable for standardization if the units are interdependent.

Critical rationalism emphasizes scrutiny and learning from errors as essential for advancing knowledge and human problem-solving. Learning from errors is also a major concern in error management with the aim of maximizing the positive effects of errors. Against this background, this paper explores the potential adoption of a critical rationalist perspective in error management. To this end, the paper examines whether error management is suitable for reflecting the major tenets of a critical rationalist perspective in problem-solving. The analysis reveals significant correspondences between error management and critical rationalism, including conceptualization and recognition of errors, and learning through trial and error. Moreover, the paper discusses contributions and challenges that adopting the “philosophical lens” of critical rationalism for error management may entail. The findings are relevant from various respects – in a more practical sense, for example, to take precautions that encourage organizational members’ critical attitude, to impede blaming, or to follow the concept of small losses, aligning with Popper’s piecemeal engineering. In a more conceptual sense, the discussion reveals that the innate dynamics of learning and unlearning that the critical rational perspective imposes may be a major issue not only for organizational members but also, by raising questions about the appropriate and acceptable levels of dynamics in organizations.

In situations of collective action, aligning the behavior of the collective’s members with the collective’s objectives is a prevailing issue studied in various domains of social science, including sociology and organization studies. Prior research suggests that members’ mutual trust could significantly contribute to resolving collective action problems. This paper aims to contribute to this field of research by showcasing the potential of agent-based modeling for integrating mutual trust in agents’ decision-making behavior. The paper introduces an agent-based model – using the framework of NK fitness landscapes – to study the emergence of trust and its performance effects when contracts cannot govern all task elements assigned to subordinate decision-makers in hierarchical organizations. As for the theoretical basis, the paper builds on the renowned integrated trust model of Brower, Schoorman, and Tan. The simulation results underpin that mutual trust among agents can significantly contribute to coordination and, thus, enhance the organization’s performance. Moreover, the results suggest an interesting “leveraging effect” of additional pieces of information on parties’ trust and performance due to the mutual trust they induce, though moderated by the complexity of interactions. The results also indicate that increasing the probability of detecting the not-contracted performance affects the emerging trust of contracting parties asymmetrically.

Most agent-based models in the organizational sciences employ some variant of hill-climbing algorithms for representing human decision-making behavior. However, experimental research suggests that hill-climbing might not appropriately represent managerial behavior, while satisficing is an empirically relevant representation. Against this background, this paper takes a step forward to explore the impact of the algorithmic representation of decision-making behavior on the results of agent-based models, especially for emerging macro-patterns. Based on the framework of NK-fitness landscapes, the paper employs an agent-based model of rudimentary organizations for distributed decision-making. The results suggest that switching from hill-climbing to satisficing shifts the trade-off between "stability and enhancement of search" to the latter. Moreover, for the macro-pattern "performance declines with increasing complexity" as emerging from hill-climbing, the simulation experiments reveal mixed observations: Not only is satisficing considerably more sensitive to intra-organizational complexity; when local satisficers strive for global performance, the macro-pattern does not universally emerge. These findings suggest that further research on how our agent-based models represent human decision-making behavior appears necessary.

PurposeThis research seeks to explore the intersection between modularity and conformity in organizational contexts. Modularity, a cornerstone of organizational design, pertains to the decomposability of tasks within an organization into subtasks with internal interdependence and external independence. Conformity, on the other hand, is the adjustment of an individual’s behavior to match that of others, often driven by a desire to adhere to social norms.Design/methodology/approachWe employ agent-based modeling and simulation as a technique to model organizations as complex systems. This approach allows us to delve into the effects of modularity in organizational structures on organizational performance, with a particular emphasis on the role of conformity in this relationship. We treat conformity as exogenously given, which allows us to focus on its effects rather than its emergence.FindingsThe results demonstrate that a concentration of interdependent tasks within fewer departments can boost overall performance. Conformity decreases performance in all organizational structures except for cases when the departments work on highly similar tasks. This decline in performance can also explain why functional organizational structures are still being used in practice even though they are less modular than divisional structures — they feature lower levels of conformity and, thus, face smaller decline. Finally, we find that in highly complex settings, organizational performance can, surprisingly, be improved as complexity within departments increases.Originality/valueTo the best of our knowledge, this study is the first to explore the modularity in organizational structures in presence of conformity. Distinctively, we adapt the NKCS model from evolutionary biology to our study, and perform an exhaustive analysis by examining all possible combinations of parameters that refer to the task allocation within organizations. We thereby contribute a unique perspective to the discourse on organizational theory and behavior.

Multi-unit organizations are a form of organizations where the geographically dispersed units provide similar products or services in different markets. Deciding on an appropriate level of centralization in such organizations presents a unique challenge. One the one hand the organizations want to maintain a consistent brand identity in all units through centralized control, but on the other hand, they want to provide the units with sufficient autonomy to respond to the challenges they face locally. Traditionally, this challenge was perceived to require a trade-off between performance and organizational synchrony, with performance demanding more decentralization and synchrony requiring more centralized control. However, our research explores how organizations can potentially resolve this trade-off by promoting norms for knowledge-sharing and setting up the right communication channels, relying on the unit managers' intrinsic tendency to conform to the behavior of their peers. We build an agent-based model of an organization with multiple interdependent units facing highly similar task environments to investigate how unit managers' ability to communicate, share knowledge, and conform to peer practices might influence organizational dynamics. We find that, under specific communication network structures, increased decentralization can enhance both performance and organizational synchrony without sacrificing one or the other. Furthermore, we discover that centralization might still be preferable for synchrony if the units are interdependent.

Karl R. Popper translated the core idea of critical rationalism to a managerial context when advocating the searching for and learning from mistakes as part of professional ethics. This paper builds on the conjecture that adopting a critical rationalist perspective contributes to advancing management control toward empirical foundation, scrutiny, and continuous learning. Based on the work of critical rationalist Hans Albert, three core components of results controls – namely cause-effect relations, provision of incentives, and accountability – are analyzed for potential obstacles for active testing and learning from mistakes. The impediments identified can be condensed into three types: cognitive limitations, self-interested behavior, and structural properties of results controls, including inherent assumptions about managerial behavior and situational conditions. The findings contribute to management control in various respects – from a more practical side toward means for counteracting the obstacles against a critical rational approach; conceptually, findings indicate so far underdeveloped issues in management control, such as learning on incentivization. Moreover, the analysis reveals that the critical rationalist approach may pose particular challenges for management control and organizational members, as it inherently may impose frequent changes and uncertainty.

Abstract Management control aims at conforming the behaviour of organizational members to organizational goals. To this aim, management control employs a multitude of control mechanisms—from performance measurements and incentive schemes to group norms and shared beliefs. This chapter focuses on the potential of agent-based modelling and simulation for theory building in management control. A core argument of the chapter is that the agent-based paradigm structurally corresponds to management control. The discussions provide examples of agent-based models contributing to emerging research topics in the field. The models are differentiated according to how they capture control mechanisms (i.e., as exogenously given and endogenously evolving). Models with exogenously given control mechanisms primarily focus on the effects and effectiveness of single mechanisms or combinations thereof, potentially differentiating for environmental contingencies. Agent-based models with endogenously evolving control mechanisms allow capturing the emergence of management control systems and the adaptation to potentially turbulent environments. The chapter discusses the promises and challenges of agent-based modelling for the domain and derives avenues for future research.

Generative artificial intelligence has the potential to both exacerbate and ameliorate existing socioeconomic inequalities. In this article, we provide a state-of-the-art interdisciplinary overview of the potential impacts of generative AI on (mis)information and three information-intensive domains: work, education, and healthcare. Our goal is to highlight how generative AI could worsen existing inequalities while illuminating how AI may help mitigate pervasive social problems. In the information domain, generative AI can democratize content creation and access, but may dramatically expand the production and proliferation of misinformation. In the workplace, it can boost productivity and create new jobs, but the benefits will likely be distributed unevenly. In education, it offers personalized learning, but may widen the digital divide. In healthcare, it might improve diagnostics and accessibility, but could deepen pre-existing inequalities. In each section we cover a specific topic, evaluate existing research, identify critical gaps, and recommend research directions, including explicit trade-offs that complicate the derivation of a priori hypotheses. We conclude with a section highlighting the role of policymaking to maximize generative AI’s potential to reduce inequalities while mitigating its harmful effects. We discuss strengths and weaknesses of existing policy frameworks in the European Union, the United States, and the United Kingdom, observing that each fails to fully confront the socioeconomic challenges we have identified. We propose several concrete policies that could promote shared prosperity through the advancement of generative AI. This article emphasizes the need for interdisciplinary collaborations to understand and address the complex challenges of generative AI.

Many agent-based models of human decision-making in organizations employ representations and algorithms comprising decision-makers’ aspirations. However, aspiration levels usually do not receive much attention in the modeling efforts, nor is agent-based modeling employed to understand better the effects and emergence of aspiration levels in decision-making. This paper elaborates on the relevance of aspiration levels in agent-based models using the widely used hill-climbing algorithms and reinforcement learning as examples. The paper provides a framework for the modeler’s multi-faceted design choices when capturing aspiration levels for decision-making with a particular focus on organizational contexts. The framework builds on the ODD + D protocol, which has been proposed explicitly for agent-based models with human decision-makers. The framework also allows deriving potential contributions of the agent-based modeling approach to understanding the effects of aspiration levels in organizations. These may, for example, include the dynamic interactions between individual and organizational aspirations, the adaptation to environmental changes, or the relevance of decision-makers’ cognitive capabilities.

Abstract This chapter examines the role of agent-based modelling and simulation in advancing theories and addressing operational and design issues within the realm of management science. We specifically explore the potential contributions of agent-based modelling and simulation to abductive and deductive reasoning. Through a theoretical lens, we argue that agent-based modelling and simulation is useful in exploring complex dynamics, and it thereby emerges as a potent tool for generating, refining, and selecting hypotheses. It facilitates a deeper understanding of the behaviours within complex adaptive systems—ranging from organizations to networks—thereby enabling the formulation of causally plausible explanations for observed phenomena. This capability is particularly crucial in environments where traditional analytical methods fall short due to inherent complexities. Furthermore, we discuss the role of agent-based modelling and simulation in operationalizing hypotheses for empirical testing, emphasizing its utility in translating theoretical concepts into testable models and identifying the boundary conditions under which these models hold true. In the context of operational and design issues, we argue that agent-based modelling and simulation supports the exploration of means–ends relations and the operationalization of design choices, thereby contributing to the development of innovative solutions and potentially better decision-making within organizations. Thereby, the method substantially contributes to bridging the gap between theory and practice within the field of management science, presenting it as an essential approach for navigating the complexities of contemporary management challenges.

Abstract In recent decades, management science has witnessed increasing interest in the reciprocal mechanisms between micro-level behaviours and macro-level patterns. At the same time, there are calls to bridge the field’s multidimensional divide in subdomains, theoretical foundations, or research methods that often address either the micro or macro level. The introductory chapter of the Handbook of Agent-based Computational Management Science outlines how the agent-based paradigm contributes to bridging the micro–macro divide in management science. For this, we build on the framework of causal mechanisms of Coleman’s boat. The dynamic variant of the framework, proposed in the philosophy of social science, allows for capturing the reciprocal connections between the micro and macro levels and translates into the agent-based paradigm and its generative approach. For the novice in agent-based modelling and simulation, we briefly outline some more ‘technical’ aspects. Finally, we provide a brief overview of the chapters in the handbook, organized into four parts as there are (1) the microfoundations of agent-based models in management, agent-based modelling and simulation for (2) theory building in management and (3) for operational issues in management, followed by some overarching, and (4) reflections and extensions.

Models of economic decision makers often include idealized assumptions, such as rationality, perfect foresight, and access to all relevant pieces of information. These assumptions often assure the models’ internal validity, but, at the same time, might limit the models’ power to explain empirical phenomena. This paper is particularly concerned with the model of the hidden action problem, which proposes an optimal performancebased sharing rule for situations in which a principal assigns a task to an agent, and the action taken to carry out this task is not observable by the principal. We follow the agentization approach and introduce an agent-based version of the hidden action problem, in which some of the idealized assumptions about the principal and the agent are relaxed so that they only have limited information access, are endowed with the ability to gain information, and store it in and retrieve it from their (limited) memory. We follow an evolutionary approach and analyze how the principal’s and the agent’s decisions affect the sharing rule, task performance, and their utility over time. The results indicate that the optimal sharing rule does not emerge. The principal’s utility is relatively robust to variations in intelligence, while the agent’s utility is highly sensitive to limitations in intelligence. The principal’s behavior appears to be driven by opportunism, as she withholds a premium from the agent to assure the optimal utility for herself.

Conceptual metaphor theory provides linguistic evidence for the assumption that the human conceptual system is largely metaphorical, pervasively affecting how individuals comprehend abstract concepts. Against this background, conceptual metaphors in the field of management control are of interest as they may affect how scholars, students, and practitioners comprehend management control. This study elaborates on conceptual metaphors in management control. In particular, it focuses on the analysis of those parts of prominent textbooks in the field which provide a general introduction to results controls employing established methods in linguistic metaphor studies. The discourse analysis reveals that, for results controls, the MACHINE metaphor and the PEOPLE metaphor are particularly prominent. These findings are discussed with respect to implicit assumptions and the broader context of organizational metaphors. Avenues for future research on conceptual metaphors in management control are proposed.

Multi-unit organizations such as retail chains are interested in the diffusion of best practices throughout all divisions. However, the strict guidelines or incentive schemes may not always be effective in promoting the replication of a practice. In this paper we analyze how the individual belief systems, namely the desire of individuals to conform, may be used to spread knowledge between departments. We develop an agent-based simulation of an organization with different network structures between divisions through which the knowledge is shared, and observe the resulting synchrony. We find that the effect of network structures on the diffusion of knowledge depends on the interdependencies between divisions, and that peer-to-peer exchange of information is more effective in reaching synchrony than unilateral sharing of knowledge from one division. Moreover, we find that centralized network structures lead to lower performance in organizations.

It has long been recognised that the identification of organizational members with their organiza-tion could mitigate collective action problems. One domain in economics and management addressing these fundamental problems is organizational control, as it focuses on aligning individuals' behavior with their or-ganization's objectives. This paper studies organizational identification in interaction with major traits of the organizational context, including the principal means of organizational control. The study uses an agent-based simulation based on the framework NK fitness landscapes. In the model, individuals' organizational identifica-tion emerges endogenously and causes certain effects in the organizations, which in turn induce feedback on identification. The model controls for different activation mechanisms for individuals' organizational identity and for different organizational contexts in terms of task complexity and the prevailing coordination mecha-nism. The results suggest that the activation mechanisms subtly interfere with task complexity and organiza-tional controls. Organizational identification was robustly beneficial across a wide range of task environments when the activation mode corresponds to organizational mission orientation and results controls. Moreover, organizational identification appears particularly relevant when the action controls grant some autonomy to subordinate decision-makers.

Limited memory of decision-makers is often neglected in economic models, although it is reasonable to assume that it significantly influences the models’ outcomes. The hidden-action model introduced by Holmström also includes this assumption. In delegation relationships between a principal and an agent, this model provides the optimal sharing rule for the outcome that optimizes both parties’ utilities. This paper introduces an agent-based model of the hidden-action problem that includes limitations in the cognitive capacity of contracting parties. Our analysis mainly focuses on the sensitivity of the principal’s and the agent’s utilities to the relaxed assumptions. The results indicate that the agent’s utility drops with limitations in the principal’s cognitive capacity. Also, we find that the agent’s cognitive capacity limitations affect neither his nor the principal’s utility. Thus, the agent bears all adverse effects resulting from limitations in cognitive capacity.

Diversity in teams has become an important societal and economic issue which is studied in various scientific domains. In organizational sciences, particularly empirical research methods prevail. This paper proposes to explore agent-based computational economics as a research approach for workforce diversity more intensely due to its inherent properties like capturing heterogeneous interacting agents. For highlighting this, this paper presents an agent-based computational model based on the framework of NK fitness landscapes. In the simulations, artificial organizations search for superior levels of organizational performance with search being delegated to several and potentially diverse decision-making agents. The experiments control for the level of task complexity and reflects four different attributes of workplace diversity among agents: cognitive capabilities to (i) generate and (ii) evaluate new solutions, (iii) effort efficiency and (iv) commitment to the overall organizational objective. The results suggest that the effects of workforce diversity differ across task complexity and attributes of diversity. Diversity of commitment has the strongest impact which results from interactions among local maximizers and agents seeking to globally maximize with only local means. Moreover, the results point to nonlinear effects of multi-attributive diversity on organizational performance.