Academic journal article International Journal of Business Studies

Defending against Turbulent Conditions: Results from an Agent-Based Simulation

Academic journal article International Journal of Business Studies

Defending against Turbulent Conditions: Results from an Agent-Based Simulation

Article excerpt

We use an agent-based computer simulation to examine the impact of organisational strategy on enterprise performance within a turbulent operating environment. The simulation is designed to model consumers and four types of small- and medium-sized enterprises that behave with different strategies (Defender, Analyser, Prospector and Static). These agent-based organisations operate with different behavioural characteristics when making strategic decisions about the number of products to stock and which products to offer for sale. We present the results of experiments conducted using a series of simulations. Outcomes from these simulations indicate that organisations that employ Defender strategies perform best in a turbulent operating environment.

Keywords: agent-based modelling, strategic orientation, turbulent environment, small-medium sized enterprises, Miles and Snow typology


The underlying motivation of our work is to understand the behaviour of small- to medium-sized enterprises operating under various environmental conditions. The strategic decisions of these organisations impact on both their resource needs and performance. Indeed their organisational strategy may be especially critical in complex and unpredictable operating environments with shifting consumer demand. Understanding the impact of different strategies in different market conditions is complex as environmental conditions tend to emerge in a dynamic way from an interaction of consumers with changing demands and enterprises with different strategic orientations.

Computer simulations, such as our agent-based model, provide a pragmatic approach to modelling such complex behaviour. While similar to traditional experimental designs, a computer simulation uses an abstract model of reality to conduct experiments rather than real world subjects (Hunter & Naylor, 1970; Achterkamp & Imhof, 1999). Computer simulation has indeed been described as a "third way of doing science" (Casti, 1997). This approach is typically undertaken to improve understanding of a phenomena, to predict future behaviour and to conduct experiments that cannot, for some reason, be carried out on the real world system (Davidsson, 2002). For example, Zott (2003) developed a computer simulation to provide insight into the trajectories of evolutionary change engendered by dynamic capability in organisations. Others such as Phelan (2002) have employed simulations to provide insights into the performance levels of organisations with differential cognitive capacities. More specifically, agent-based approaches have proved useful in exploring market scenarios (Csik, 2003; Garifullin et al., 2007).

We have implemented an agent-based simulation that can be configured to create stable, trending or turbulent operating environments (Figure 1). These figures show characteristic consumer demand for 12 different products when the simulation is configured for a stable, trending and turbulent environment. Note that this behaviour is only characteristic as each simulation run will produce unique results. These three different environmental conditions have been used previously to model the adaptive behaviour of organisations (Friesen & Miller, 1986; Blackmore, 2006). In this work we focus on turbulent environments. A turbulent environment is considered to be dynamic and discontinuous, characterised by abrupt shifts. In this scenario consumer demand for products fluctuates over the course of the simulation.

The principal agent types used in our simulation represent consumers and enterprises. The consumer agents are driven by a simple goal of seeking a product from the enterprise agents. We then model environmental conditions by modifying consumer behaviour in terms of what products they are seeking. Turbulence, for example, is modelled by introducing larger degrees of randomness into consumer goal setting, thus making consumer behaviour less predictable. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.