Solving Last-Mile Deliveries for Dairy Products Using a Biased Randomization-Based Spreadsheet. A Case Study

Abstract

Abstract During recent years, Last-mile deliveries (LMD) have become relevant due to its application in e-commerce, urban logistics, and food delivery among others. This work addresses an LMD denoted as a Vehicle Routing Problem with time windows (VRPTW) and aims to minimize total time of the distribution process (i.e., makespan). The LMD is an NP-Hard problem that refers to the delivery of goods from a consolidation center to a destination. For solving the problem, a spreadsheet-based solution that employs a multi-start algorithm based on the biased-randomized version of the nearest neighbor heuristic is introduced. Real historical data of last-mile deliveries for dairy products in Bogotá (Colombia) was considered for evaluating our proposed method. Computational experiments are carried out to show the competitiveness of our method in terms of makespan, number of vehicles, average vehicle occupancy, average load and costs. Some insights for future works are also provided.