Solving the woody supply chain for Ireland's expanding biomass sector: A case study

For biomass energy systems to be cost-competitive, biomass must be delivered at the lowest possible costs. Truck transportation is not only the most common method used for supplying wood biomass, but it also constitutes a high proportion of the wood supply chain. Truck scheduling is part of the operational planning; the aim is to find a set of daily minimal cost routes for each truck in a fleet such that all customer demands are satisfied on time without exceeding the supplies available at the supply points.This chapter presents an approach for solving the log truck scheduling problem under Irish scenarios. The problem consisted of a heterogeneous fleet of 33 trucks based on 14 depots and carrying woodchips and logs. The problem consisted of 89 transportation tasks or deliveries to satisfy the energy demand from three peat power stations and the log demand of two panel board mills. The solution approach presented is based on a standard simulated annealing (SA) procedure. The problem was solved with a tool developed for the Microsoft Windows platform. It was programmed in Visual C++ using an object-oriented design. It includes the SA heuristics as an optimization engine in combination with a deterministic discrete event simulation to emulate the movement of trucks throughout the day.The model exported the results on a series of tables presenting the metrics of the truck performance and the daily schedule. A sensitivity analysis was conducted to determine the effect of truck configuration, driving route selection, and how the truck's payload is influenced by the biomass moisture content (MC) and loading and unloading times. Results showed that using five axle trucks reduced the total biomass supplied by 12% in comparison with 6-axled trucks, thus increasing the cost per ton delivered. Trucks choosing routes with the shortest distance decreased the overall transportation costs by 2.2%. However, these routes implied higher driving times (10.71%), which translated into 14.8% fewer deliveries and therefore the cost per ton increasing by 12.93%. The MC of wood did not affect the overall performance of how the trucks were scheduled, but it had an effect on the truck's productivity as it reduced the payloads and therefore the total tons supplied. The drier wood means higher energy content, increasing the payment per dry ton. Modeling the increase in loading times by 10% and 20% resulted in a reduction of the number of truckloads by 5.36% and 12.5%, respectively. In general, driving time and MC were the major factors affecting wood biomass delivery and the transportation costs per unit. © 2016 Elsevier Ltd. All rights reserved