Three Essays on the Economic Impacts of Harmful Algal Blooms
Transportation and logistics of recyclable materials are major cost drivers within recycling systems. In low and medium income countries, the collection costs of municipal solid waste (MSW) respectively can take up to 80-90% and 50-80% of the municipal solid waste management budget. Reducing these costs can significantly improve the viability of recycling systems. A promising approach for cost reduction in this context is improved routing of refuse vehicles, which can be done by developing and using advanced mathematical optimization algorithms and data information systems. Within the optimization research field, vehicle routing is addressed within the capacitated vehicle routing problem (CVRP). The objective is to minimize the total cost of a fleet of vehicles with a designated capacity. The CVRP is stated as an NP-Hard problem in combinatorial optimization problems meaning there is no known algorithm to solve these types of problems within polynomial time. Various heuristics are used to solve the CVRP and this study aims to use a genetic algorithm. The genetic algorithm provides feasible results to the CVRP within reasonable time. Applying the CVRP to a real-world instance requires road network distances compared to euclidean distances. In this thesis, a CVRP is solved using data processed in a geographical information system (GIS), existing local government databases, a routing machine, and a genetic algorithm to estimate recycling costs and emissions for municipalities in the state of Rhode Island.