Simple and practical optimization approach based to solve a truck load and delivery problem at long haul distances with heterogenous products

Abstract

This paper proposes an optimization based approach for solving the logistic processes of deliveries scheduling and product accommodation during loading with a heterogeneous fleet of vehicles. The approach focuses on the case of products with “low density values” and high heterogeneous volume and weight, and with traveling large distances to different zones, in which transportation costs constitute a important proportion of total logistic costs. The proposed approach consists of a two-phase strategy: The first uses a “Cutting Stock Problem” formulation to define utilization areas inside trucks assigned to each product family. This task is achieved by minimizing the long-haul transportation costs as a function of the vehicle size, considering a set of predefined solutions for feasible and efficient loading obtained as a result of the accumulated experience. The second phase consists of Bin Packing Problem version with a known number of bins, which were previously determined in the first phase of the approach. In this phase, different orders from a set of customers are assigned to each truck by obeying the predefined utilization areas per product category obtained in the first phase while minimizing the number of visits of each truck. The results show that the model addresses the analyzed problem in an efficient manner, which is reflected in reasonable resolution times and costs from a practical implementation perspective. Additionally, it is observed that long-haul delivery costs and vehicle utilization tend to improve with the increase of the utilized number of patterns even when the execution time is incremented.