Modeling forest sector structural evolution with the experience-weighted-attraction-learning (EWA-lite) algorithm

Abstract

The conventional economic forest sector models have limited spatial applications. In this chapter, we present an agent-based forest sector modeling framework (Cambium) that enhances spatial relevance. The model enables the study of industry interactions and strategic decision making in an environment that is characterized by continuously changing conditions in both the underlying resource inventory and finished product markets. In this model, decision processes are modeled using an implementation of the self-tuning experience weighted attraction learning algorithm (EWA-Lite). This algorithm allows agents to adjust their learning behavior along a continuum between reinforcement learning and belief learning depending on the perceived stability of their environment. The use of three distinct investment strategies (capacity expansion, process innovation, and sustainment) was found to be sufficient for achieving agent differentiation and achieving dynamic industry structure equilibrium. The number and relative size of competitors is determined by repeated agent interactions, and can be interpreted as an emergent property of the inventory, industry, and market system.