An IoT-based Smart and Sustainable Waste Management System for a Norwegian Municipality

Abstract

A solid waste management system (SWMS) is the backbone of a smart and sustainable city (SSC). The SWMS includes all the factors related to waste such as collection, sorting, transportation, and recycling. The core processes of waste collection and transportation remain the same for decades. With the help of IoT-based data driven and data transfer technologies, cameras, actuators and sensors, the waste collection process can be improved and modified. The smart waste collection triggers a transition from fixed collection to on-demand collection assigned by optimization algorithms and smart web-applications. In this thesis, a Norwegian municipality is investigated as a case study. An overview of the current practices and infrastructure is presented and the currently faced issues are addressed. A conceptual model of a smart and sustainable waste management system along with optimized waste collection and transportation system is proposed to achieve the key performance indicators (KPIs) presented in EU agenda of United for Smart and Sustainable Cities (U4SSC) (ITU-T (2016)). The proposed conceptual model illustrates the role of each stakeholder in the waste management system’s development. An IoT-based smart waste bin is proposed to detect the waste material and the waste volume in a bin. It addresses the battery life problem in IoT-based devices. A smart waste bin application is presented to directly connect the citizens and services providers for the development of a solid waste management system (SWMS). Regarding waste collection and transportation, various optimization techniques are proposed such as multi-objective traveling salesman problem (MOO-TSP), a conventional vehicle routing problem (VRP), and capacitated vehicle routing problem (CVRP). The proposed optimized solutions calculate routes for a various number of vehicles under different scenarios having specified constraints such as minimum traveling distance, mini-mum traveling time, vehicle capacity, waste volume in the bin and waste collection on demand. These solutions are cost-effective and time-efficient. In MOO-TSP, the shortest possible route is calculated with objectives such as minimum traveling distance and minimum traveling time. The calculated route takes one vehicle at a time into consideration. The proposed optimization model is 34% cost saving as compared to the current practices. In VRP, the shortest possible routes are calculated for various vehicles. The routes are calculated for different constraints such as minimum traveling time and minimum traveling distance. Furthermore, the CVRP adds several more constraints for route calculation such as capacity of a vehicle, volume of waste in each bin. Pertaining to the visualization of optimized routes on a map. Google maps platform is used to visualize the calculated routes from a depot site to several bins and back to the depot site. For optimizing routes, the greedy descent algorithm is used to calculate the nearest possible neighbour in a routing problem. The routes on Google maps are differentiated by different colors for each vehicle. The proposed optimal solutions for routing and vehicle allocation lead to a significant reduction in the overall cost of a waste management system. The results are obtained from the data from the waste bins placed by the service providers in the investigated municipality. To calculate the optimal distance and optimal time for waste calculation, the Google Distance Matrix API is used. The proposed solutions prove that IoT-based SWMS is the best replacement of current practices in order to achieve sustainable development goals (SDGs). This thesis achieves the SDG 11.6 and supports SDG 12.4, SDG 12.5. Where SDG 11 targets to make cities and human settlements inclusive, safe, resilient, and sustainable, and SDG 12 ensures sustainable consumption and production patterns (SGDs (2020)).

A solid waste management system (SWMS) is the backbone of a smart and sustainable city (SSC). The SWMS includes all the factors related to waste such as collection, sorting, transportation, and recycling. The core processes of waste collection and transportation remain the same for decades. With the help of IoT-based data driven and data transfer technologies, cameras, actuators and sensors, the waste collection process can be improved and modified. The smart waste collection triggers a transition from fixed collection to on-demand collection assigned by optimization algorithms and smart web-applications. In this thesis, a Norwegian municipality is investigated as a case study. An overview of the current practices and infrastructure is presented and the currently faced issues are addressed. A conceptual model of a smart and sustainable waste management system along with optimized waste collection and transportation system is proposed to achieve the key performance indicators (KPIs) presented in EU agenda of United for Smart and Sustainable Cities (U4SSC) (ITU-T (2016)). The proposed conceptual model illustrates the role of each stakeholder in the waste management system’s development. An IoT-based smart waste bin is proposed to detect the waste material and the waste volume in a bin. It addresses the battery life problem in IoT-based devices. A smart waste bin application is presented to directly connect the citizens and services providers for the development of a solid waste management system (SWMS). Regarding waste collection and transportation, various optimization techniques are proposed such as multi-objective traveling salesman problem (MOO-TSP), a conventional vehicle routing problem (VRP), and capacitated vehicle routing problem (CVRP). The proposed optimized solutions calculate routes for a various number of vehicles under different scenarios having specified constraints such as minimum traveling distance, mini-mum traveling time, vehicle capacity, waste volume in the bin and waste collection on demand. These solutions are cost-effective and time-efficient. In MOO-TSP, the shortest possible route is calculated with objectives such as minimum traveling distance and minimum traveling time. The calculated route takes one vehicle at a time into consideration. The proposed optimization model is 34% cost saving as compared to the current practices. In VRP, the shortest possible routes are calculated for various vehicles. The routes are calculated for different constraints such as minimum traveling time and minimum traveling distance. Furthermore, the CVRP adds several more constraints for route calculation such as capacity of a vehicle, volume of waste in each bin. Pertaining to the visualization of optimized routes on a map. Google maps platform is used to visualize the calculated routes from a depot site to several bins and back to the depot site. For optimizing routes, the greedy descent algorithm is used to calculate the nearest possible neighbour in a routing problem. The routes on Google maps are differentiated by different colors for each vehicle. The proposed optimal solutions for routing and vehicle allocation lead to a significant reduction in the overall cost of a waste management system. The results are obtained from the data from the waste bins placed by the service providers in the investigated municipality. To calculate the optimal distance and optimal time for waste calculation, the Google Distance Matrix API is used. The proposed solutions prove that IoT-based SWMS is the best replacement of current practices in order to achieve sustainable development goals (SDGs). This thesis achieves the SDG 11.6 and supports SDG 12.4, SDG 12.5. Where SDG 11 targets to make cities and human settlements inclusive, safe, resilient, and sustainable, and SDG 12 ensures sustainable consumption and production patterns (SGDs (2020)).