Waste Collection Route Optimization for the City of Oshawa

Generating household waste is an unavoidable part of daily life. Of course, it is ideal to generate as little waste as possible, but there will always be a need for curb side municipal solid waste collection to properly dispose of the generated waste within urban centres. Due to this unavoidability, it is crucial that waste collection vehicles are used in an efficient manner which considers the economic, social, and environmental effects of daily use. This study introduces a means of evaluating the most efficient path for a service route in the form of geographic information system (GIS) data utilizing a modified version of Hierholzer’s algorithm in conjunction with Dijkstra’s algorithm. The proposed algorithm includes proper interpretation of serviceable/travel roads, collection/travel fuel consumption coefficients, and dynamic speed which is based on the speed limit of the roads within the network and their respective service status. In addition to this, it is common that waste collection is done with a fleet of vehicles, many of which may be covering more than one collection route a day. Using the statistics generated for each collection route, normalized objective function values can be computed which are made up of the fuel consumption, collection time, travel time, and total distance travelled. A greedy combinatorial optimization algorithm can then be used to calculate the best-balanced collection area pairs that consist of multiple routes. A case study of the proposed algorithm was conducted in Oshawa, Ontario, consisting of many individual collection routes that can be grouped into 10 collection areas, 2 for each day of the working week. The results of this case study will be used to support the claims that the workload balance algorithm can improve the collection area pairs within the city.