Content area
Full Text
1. Introduction
Agent-based simulation has been utilized in various fields of research and the approach has gained increasing interest in operations management and supply chain management (Dyke Parunak et al., 1998; Giannakis and Louis, 2011; Amini et al., 2012). Most of the simulation studies are still using discrete events (DE) or system dynamic simulation. Traditionally flow of goods or people is handled by using numerical variables or objects containing property information. Agent-based simulation models bring some new possibilities to handle large amount of actors – agents – which may have complicated behavior and can interact with each other.
Agent-based simulation is applied when there are various participants in system who acts independently, interact with each other, react to system changes, and their total activity is nonlinear and not derived from summation of each individuals’ behavior (Ge et al., 2015). According to Vriend (2000) there are two types of agents learning; individual and population levels. In individual level, agents learn from its own experience, but in population level, learning occurs from other agents.
Modeling and development approaches developed for operations management are used in service supply chains (Sengupta et al., 2006). Generally, service supply chains refers to planning and management of activities from support functions to end-users (Voudouris et al., 2007). Material flow may be part of the service supply chain but not the main consideration as in supply chain management. Service supply chains have been studied in several fields to build frameworks (Baltacioglu et al., 2007) and modeling system-wide effects such as demand amplification (Akkermans and Vos, 2003).
Process improvement and key performance indicators (KPI) for process evaluation share many similarities between traditional manufacturing operations and service operations such as customer service, maintenance and banking services. There are differences too: as services are intangible by nature and require interaction with people, not only physical material, and as evaluation of service quality is based on individual assessment of gap between expectations and reality, management of service has its own characteristics.
This paper illustrates possibilities of using agent-based simulation in service supply chains by showing a generic example of healthcare location problem. The model features the following aspects enabled by agent-based approach:
-
use of geographical data for customer demand pattern generation;