Optimization model with multiple constraints using genetic algorithm method for cargo arrangement problem

• Main Author: Jie, Zhou
• Format: Thesis
• Language: English
• Published: 2021
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utm.my/id/eprint/99511/1/ZhouJieMKE2021.pdf; http://eprints.utm.my/id/eprint/99511/; http://dms.library.utm.my:8080/vital/access/manager/Repository/vital:149882
• Institution: Universiti Teknologi Malaysia

Efficient arrangement of cargo in logistics is crucial in minimizing the operational cost and it can be a complex task as it involves multiple constraints like cargo with various volumes and weights. Therefore, manual cargo arrangement is challenging, especially when the types of cargo and numbers of customer’s increase. Cargo arrangement is categorized as a problem that involves mathematical models and efficient optimization algorithms. This project proposes a multi-objective, multiconstraint mathematical model for the three-dimensional optimization problem (3- DOP), with constraints such as packaging volume, weight and quantity of different cargo types. The requirements and characteristics of the container are also considered in establishing the proposed model to achieve the loading optimization objectives of maximizing the utilization and capacity of container space. This project uses genetic algorithm (GA) as the global search properties to obtain the optimal solution. The proposed model comprised of an objective function and a set of constraints in cargo loading such as weight, rotation, overlapping and stacking constraints. The real coded methods of GA, optimal preservation strategy and Pareto front are introduced. Subsequently, a GA is developed using MATLAB software. The cargo sizes of different transport companies are used as test samples. The maximum space utilization is achieved up to 77.59%, and the weight maximum is 58.12%. Nevertheless, it is observed that an increase in the number of constraints has a significant effect on the optimization. In short, the effectiveness of the proposed optimization algorithm is verified.