When a series of operations are linked into a long chain of events, it is often difficult to see which operation is having the most impact on cost-effectiveness. Computer-assisted decision making tools are useful to select the most cost-effective logistics system. Recent studies [30-33] compared a number of different scenarios to determine the best unit operation options and identify bottlenecks. Mathematical programming approaches have been used to develop optimization models that are applicable to a variety of cases studies [34-37]. More recently, focus has shifted to simulation models using object-oriented programming [38] or discrete event simulation [39, 40]. The object-oriented approach [22, 41-43] simplifies scenario building, particularly related to various equipment options available for the same operation, since data pertaining to different options are stored in a standard object-oriented format.

The biomass logistics subject area can be divided into two types of modeling approaches depending on the environment encountered: stochastic or deterministic, and integer or continuous. In the areas of stochastic and deterministic environment, ISBAL and BioFeed capture the stochastic biomass logistics issues such as variability in processing as a result of weather, time, equipment breakdowns, etc. Additional work in this area can be found in [34; 40; 44-45]. In the deterministic environment, most work utilizes a geographic information system (GIS) interactively with optimization. This allows the GIS framework to work as a data management tool, which may call the optimization software directly. Most of the deterministic models are mixed-integer programs [37, 46-49]. One of the principles of designing an effective logistics system is starting with the feedstock resources distribution, and the simplest method is to utilize a GIS framework for data management.

Continuous models assume that all the land within a region is utilized for biomass production [50-52]. This approach uses average haul distances and cost. Hence, the solution obtained is difficult to implement due to lack of detail, as opposed to the solution obtained using an integer model formulation, where each production field is considered as a specific entity with specific costs, thereby resulting in precise decisions for implementation.