Computer simulation is a prominent method in organizational studies and strategic management. While there are many uses for computer simulation, most academics in the fields of strategic management and organizational studies have used computer simulation to understand how organizations or firms operate. More recently, however, researchers have also started to apply computer simulation to understand organizational behaviour at a more micro-level, focusing on individual and interpersonal cognition and behavior such as team working. While the strategy researchers have tended to focus on testing theories of firm performance, many organizational theorists are focused on more descriptive theories, the one uniting theme has been the use of computational models to either verify or extend theories. It is perhaps no accident that those researchers using computational simulation have been inspired by ideas from biological modeling, ecology, theoretical physics and thermodynamics, chaos theory, complexity theory and organization studies since these methods have also been fruitfully used in those areas.
Basic distinctions/definitions
Researchers studying organizations and firms using computer simulations utilize a variety of basic distinctions and definitions that are common in computational science
Agent-based vs Equation-based: agent-based models unfold according to the interactions of relatively simple actions, while equation-based models unfold numerically based on a variety of dynamic or steady-state equations
Model: simplified versions of the real world that contain only essential elements of theoretical interest
Complexity of the model: the number of conceptual parts in the model and the connections between those parts
Deterministic vs. Stochastic: deterministic models unfold exactly as specified by some pre-specified logic, while stochastic models depend on a variety of draws from probability distributions
Optimizing vs. Descriptive: models with actors that either seek optimums or do not
Methodological approaches
There are a variety of different methodological approaches in the area of computational simulation. These include but are not limited to the following.
Agent-based models: computational models investigating the interaction of multiple agents
Cellular automata: models exploring multiple actors in physical space whose behavior is based on rules
Dynamic network models: any model representing actors and non-actor entities as connected through relational links as in dynamic network analysis
Genetic Algorithms: models of agents whose genetic information can evolve over time
Equation-based : models using equations that determine the future state of its systems
Social Network models: any model representing actors as connected through stereotypical 'ties' as in social network analysis
Stochastic Simulation: models that involve random variables or source of stochasticity
System dynamics: equation-based approach using casual-loops and stocks & flows of resources
NK modeling: actors modeled as N nodes linked through K connections that are trying to reach the peak of a fitness landscape
Early research
Early research in strategy and organizations using computational simulation concerned itself with either the macro-behavior of systems or specific organizational mechanisms. Highlights of early research included:
Cohen, March, & Olsen's Garbage Can Model of Organizational Choice modeled organizations as a set of solutions seeking problems in a rather anarchic 'garbage can'-esque organization.
March's study of Exploration and Exploitation in Organizational Learning utilized John Holland's basic explore/exploit distinction to show the value of slow learners in organizations.
Nelson & Winter's Evolutionary theory of economic change used a simulation to show that an evolutionary model could produce the same sort of GDP / productivity numbers as neo-classical rational choice theorizing.
Later research
Later research using computational simulation flowered in the 1990s and beyond. Highlights include:
Carroll & Harrison's model of organizational demography and culture
Davis, Eisenhardt & Bingham's model of organization structure in unpredictable environments
Gavetti, & Levinthal's model of cognitive and experiential search
