Hybrid evolutionary optimization for takeaway order selection and delivery path planning utilizing habit data

Abstract The last years have seen a rapid growth of the takeaway delivery market, which has provided lot jobs for deliverymen. However, increasing numbers orders and corresponding pickup service points made order selection path planning key challenging problem to In this paper, we present integrating deliverymen, objective is maximize revenue per unit time subject maximum length, overdue penalty, reward/penalty large/small number orders, high customer scoring reward. Particularly, consider uncertain ready satisfaction level, are estimated based on historical habit data stores customers using machine-learning approach. To efficiently solve problem, propose hybrid evolutionary algorithm, adapts water wave optimization (WWO) metaheuristic evolve solutions main employs tabu search route each solution. Experimental results test instances constructed real food application demonstrate performance advantages proposed algorithm compared set popular algorithms.