Smart Waste Classification And Collection System:A Bi-Objective Modeling And Optimization Approach

In the era of the “smart city”,internet of thing and information and communication technologies have become indispensable in the planning and design of modern municipal solid waste management.Simultaneously,due to more waste varieties,there are urgent calls to implement waste classification worldwide,which promotes resource recycling to achieve sustainable development.In this paper,we present a smart waste classification and collection system that is abstracted as a bi-objective mathematical programming model to optimize the waste collection problem.To implement the proposed smart waste classification and collection system effectively,we designed a novel multi-objective hybrid algorithm based on the whale optimization and genetic algorithms with an improved convergence factor and a fast,nondominated sorting method.A comparison of our algorithm with two classical multi-objective algorithms on generated test instances and on a real-world case shows that the proposed multiobjective hybrid algorithm based on the whale optimization and genetic algorithms is more effective for optimizing the established model.This paper demonstrates how the smart waste classification and collection system works and how it can help sanitation companies improve waste collection both economically and environmentally.Our work makes the following contributions:(1)We propose using smart waste classification and collection system in the context of a smart city to optimize municipal solid waste management.The system applies internet of thing and information and communication technologies to waste classification to improve sustainability,enhance pickup trucks’ workload balance,and minimize total costs,which include transportation and carbon tax costs.It also bridges a gap in the theory and practice of municipal solid waste management by applying internet of thing and information and communication technologies to waste classification and collection.(2)To optimize the system,we designed a novel multi-objective hybrid algorithm based on the whale optimization and genetic algorithms based on the whale optimization algorithm and applied crossover and mutation.The whale optimization algorithm has strong global search ability and a fast convergence speed,but it is easy to fall into local optimization prematurely,so the whale optimization algorithm is not effective for solving complex optimization problems.Our proposed multi-objective hybrid algorithm based on the whale optimization and genetic algorithms obtains better solutions by using crossover and mutation to expand search depth.It also has an improved convergence factor to enhance its early global search capabilities and increase the probability of jumping out of local optimization during the later period.Finally,it is combined with a fast non-dominated sorting method,thereby providing effective algorithm support for the smart waste classification and collection system.(3)To verify the proposed algorithm’s effectiveness,we designed test instances and conducted computational experiments to compare it with two classic multi-objective optimization algorithms: the non-dominated sorting genetic algorithm Ⅱ and the multi-objective evolutionary algorithm based on decomposition.We also tested the proposed method on a realworld case.The solutions obtained through multi-objective hybrid algorithm based on the whale optimization and genetic algorithms showed better approximation and distribution,making multi-objective hybrid algorithm based on the whale optimization and genetic algorithms an excellent choice to optimize the smart waste classification and collection system.Our analysis yields the following results.First,the smart waste classification and collection system established in this paper can effectively help sanitation enterprises save operating costs,fulfill their social responsibility to protect the environment,and improve employee satisfaction.Second,through several computational experiments and the verification on a real-world case,the proposed multi-objective hybrid algorithm based on the whale optimization and genetic algorithms is effective and superior in solving the built model.