An agent-based supply chain model for strategic analysis in foresty

• Main Author: Vahid, Saba
• Language: English
• Published: University of British Columbia 2011
• Online Access: http://hdl.handle.net/2429/39780

An agent-based forest sector model, CAMBIUM 2.0, is developed and applied to case studies of the forest industry in the coastal British Columbia (BC). By combining optimization and simulation, this model allows policy makers and managers to examine the impact of different supply chain (SC) configurations (e.g. establishing new facilities), and changing forest management policies (e.g. harvest restrictions). The forest sector structure and the state of the forest resources that develop over time are a result of autonomous agents interacting with each other while competing for available forest resources needed to manufacture forest products. The thesis is presented in four chapters. In chapter 1, I introduce SC modelling concepts and techniques and identify research objectives and methods. Chapter 2 presents and discusses the structure of the agent-based simulation model and the formulation of the facility location problem, presenting a novel algorithm for integrating the optimization problem with the simulation model. The model is applied to the case of a forest industry SC to establish a new agent. The predictions of the new agent about its profits are not strongly affected by higher levels of information about the cost structure of its competitors, while improving the accuracy of market predictions has a noticeable impact on such predictions. Chapter 3 evaluates the impact of establishing a log sort yard on the profitability of the forest products SC. Considering different market price scenarios, establishing a sort yard does not seem to benefit the forest products SC, mainly because of intense competition for timber. In Chapter 4, CAMBIUM 2.0 is used to investigate the impact of harvest policy changes on the SC performance and the timber supply sustainability. Alternative harvest priorities (e.g. harvesting stands with highest value first) and modifying the harvesting preference of the mills (i.e. harvesting a mix of high and low value stands) improves the timber supply sustainability with less negative economic impacts compared to lowering the harvest limit. The modelling framework developed in my research can be extended to address other research questions such as changing log export policies, setting stumpage prices, or encouraging replanting of economically desirable species.