Adaptive Large Neighborhood Search Algorithm for Smart Farm Picking and Distribution Integration Problem

This paper addresses an integration decision of the robot task scheduling in a picking process and the vehicle routing in a distribution process to improve the production efficiency in the context of farm-to-door mode. To address the integrated problem, an adaptive large neighborhood search (ALNS) algorithm is proposed with the objectives of minimizing the total cost occurred in the two processes. Based on the characteristics of the problem and our previous study, we employ an integer-based solution representation and generate a high-quality initial solution by using the nearest neighbor heuristic algorithm. The remove and repair operators are introduced to allow the algorithm to explore different regions of the solution space. The purpose of the simulated annealing acceptance criterion is to enhance global search capabilities by permitting the acceptance of worse solutions. Additionally, the selection probabilities of the operators are adjusted based on their performance, ensuring that more effective operators are used more frequently during the search, thereby improving both the search efficiency and the quality of the solutions. Experimental results demonstrate that the ALNS is a new state-of-the-art algorithm for the problem under consideration.