隨著網路購物的興盛，許多企業為了專注於自身品牌的提升，將貨物運輸任務交付於第三方物流進行管理。作為供應鏈上下游間的中繼站，第三方物流的訂單管理及分配對其經營效益與口碑甚為重要。本研究開發一套針對第三方物流配送最適運輸商分配演算法，透過目標式的訂定，在滿足訂單條件與需求的情況下顧及不同的效益目標。利用機器學習擅長於大量資料中習得潛在關聯的特性，以第三方物流的大量訂單資料訓練極限梯度提升模型（XGBoost），預測合適的運輸商。此外，為了提升運營績效，使用基因演算法針對運輸成本、時間與品質進行優化，並同時考量運輸商最小分配份額的限制，提供更全方位的物流解決方案。而為了加速基因演算法的迭代過程，以XGBoost的預測結果作為基因演算法之初始值，提升其運算效率。為驗證所提出之方法，本研究使用15987筆業界第三方物流真實訂單資訊進行模型驗證。實驗結果發現與傳統的Rule-based分配方法相比，XGBoost模型可取得98.1%的預測準確率及0.002秒的高效執行速度。而基因演算法透過多項目標進行運輸商優化實驗中，雖然受到訂單本身需求之限制，仍能提升約2.3%的適應度值，為第三方物流取得最大化的運營效益。

With the booming popularity of online shopping, many businesses, in order to focus on enhancing their own brand, entrust the task of goods delivery to third-party logistics for management. As a relay station in the supply chain, the order management and allocation of third-party logistics play a crucial role in its operational efficiency and reputation. In this study, an algorithm for distributing the most suitable parcel delivery company for third-party logistics allocation was proposed, which taking into account different efficiency objectives while meeting order conditions and requirements by setting objective function. The algorithm utilizes the characteristics of machine learning, which excels at discovering underlying correlations in large amounts of data. Predicting suitable parcel delivery company by training an Extreme Gradient Boosting (XGBoost) model with a large volume of order data from third-party logistics. In addition, to improve operational performance, a genetic algorithm is employed to optimize transportation costs, speed, and quality, while consider the minimum amount guarantee, to provide a more comprehensive logistics solution. To accelerate the iterative process of the genetic algorithm, the prediction results of XGBoost are used as the initial value of genetic algorithm to enhance its computational efficiency. In order to validate the proposed methods, the model was verified using 15,987 real orders from third-party logistics. The experimental results revealed that the XGBoost model achieves 98.1% prediction accuracy and 0.002 seconds efficient execution speed compared to the traditional Rule-based allocation method. And Genetic Algorithm has been able to maximize the operational efficiency of the third-party logistics by improving the fitness value approximately 2.3%, despite the limitations of the order demands.

[1] P. Prashant, G. Saji, and M. Arqum, "Trends in third party logistics–the past, the present & the future," International Journal of Logistics Research and Applications, vol. 24, no. 6, pp. 551-580, 2021.
[2] D. Werner, A. Sascha, and G. Martin, "The impact of electronic commerce on logistics service providers," International journal of physical distribution & logistics management, vol. 32, no. 3, pp. 203-222, 2002.
[3] A. Aicha, "Third-party logistics selection problem: A literature review on criteria and methods," Omega, vol. 49, pp. 69-78, 2014.
[4] X. Hu, G. Wang, X. Li et al., "Joint decision model of supplier selection and order allocation for the mass customization of logistics services," Transportation Research Part E: Logistics and Transportation Review, vol. 120, pp. 76-95, 2018.
[5] E. J. Anderson, T. Coltman, T. M. Devinney et al., "What drives the choice of a third‐party logistics provider?," Journal of Supply Chain Management, vol. 47, no. 2, pp. 97-115, 2011.
[6] T. H. Ly, S. Roh, and H. Jang, "Selection of functional logistics service providers: AHP and DEMATEL application," Korean Data Analysis Societ, vol. 23, no. 4, pp. 1517-1534, 2021.
[7] S. Islam, S. H. Amin, and L. J. Wardley, "Machine learning and optimization models for supplier selection and order allocation planning," International Journal of Production Economics, vol. 242, p. 108315, 2021.
[8] C. H. Papadimitriou and K. Steiglitz, Combinatorial optimization: algorithms and complexity. Courier Corporation, 1998.
[9] M. A. Rahman, R. Sokkalingam, M. Othman et al., "Nature-inspired metaheuristic techniques for combinatorial optimization problems: overview and recent advances," Mathematics, vol. 9, no. 20, p. 2633, 2021.
[10] M. Baghel, S. Agrawal, and S. Silakari, "Survey of metaheuristic algorithms for combinatorial optimization," International Journal of Computer Applications, vol. 58, no. 19, 2012.
[11] J. M. Weinand, K. Sörensen, P. S. .Segundo et al., "Research trends in combinatorial optimization," International Transactions in Operational Research, vol. 29, no. 2, pp. 667-705, 2022.
[12] I. Deb and R. K. Gupta, "A genetic algorithm based heuristic optimization technique for solving balanced allocation problem involving overall shipping cost minimization with restriction to the number of serving units as well as customer hubs," Results in Control and Optimization, vol. 11, p. 100227, 2023.
[13] M. Abdel-Basset, L. Abdel-Fatah, and A. K. Sangaiah, "Metaheuristic algorithms: A comprehensive review," Computational intelligence for multimedia big data on the cloud with engineering applications, pp. 185-231, 2018.
[14] D. Whitley, "A genetic algorithm tutorial," Statistics and computing, vol. 4, pp. 65-85, 1994.
[15] S. Katoch, S. S. Chauhan, and V. Kumar, "A review on genetic algorithm: past, present, and future," Multimedia tools and applications, vol. 80, pp. 8091-8126, 2021.
[16] K. Deb, A. Pratap, S. Agarwal et al., "A fast and elitist multiobjective genetic algorithm: NSGA-II," IEEE transactions on evolutionary computation, vol. 6, no. 2, pp. 182-197, 2002.
[17] W. Liu, R. Wu, Z. Liang et al., "Decision model for the customer order decoupling point considering order insertion scheduling with capacity and time constraints in logistics service supply chain," Applied Mathematical Modelling, vol. 54, pp. 112-135, 2018.
[18] M. J. Songhori, M. Tavana, A. Azadeh et al., "A supplier selection and order allocation model with multiple transportation alternatives," The International Journal of Advanced Manufacturing Technology, vol. 52, pp. 365-376, 2011.
[19] K. P. Murphy, Machine learning: a probabilistic perspective. MIT press, 2012.
[20] I. H. Sarker, "Machine learning: Algorithms, real-world applications and research directions," SN computer science, vol. 2, no. 3, p. 160, 2021.
[21] M. I. Jordan and T. M. Mitchell, "Machine learning: Trends, perspectives, and prospects," Science, vol. 349, no. 6245, pp. 255-260, 2015.
[22] L. A. Clark and D. Pregibon, "Tree-based models," in Statistical models in S: Routledge, 2017, pp. 377-419.
[23] J. Ali, R. Khan, N. Ahmad et al., "Random forests and decision trees," International Journal of Computer Science Issues (IJCSI), vol. 9, no. 5, p. 272, 2012.
[24] T. Chen and C. Guestrin, "Xgboost: A scalable tree boosting system," in Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, 2016, pp. 785-794.
[25] H. Wang, Z. Lei, X. Zhang et al., "Machine learning basics," Deep learning, pp. 98-164, 2016.
[26] J. D. Rodriguez, A. Perez, and J. A. Lozano, "Sensitivity analysis of k-fold cross validation in prediction error estimation," IEEE transactions on pattern analysis and machine intelligence, vol. 32, no. 3, pp. 569-575, 2009.
[27] J. Bergstra and Y. Bengio, "Random search for hyper-parameter optimization," Journal of machine learning research, vol. 13, no. 2, 2012.
[28] J. D. Novaković, A. Veljović, S. S. Ilić et al., "Evaluation of classification models in machine learning," Theory and Applications of Mathematics & Computer Science, vol. 7, no. 1, p. 39, 2017.
[29] M. Mitchell, An introduction to genetic algorithms. MIT press, 1998.
[30] L. Haldurai, T. Madhubala, and R. Rajalakshmi, "A study on genetic algorithm and its applications," Int. J. Comput. Sci. Eng, vol. 4, no. 10, pp. 139-143, 2016.
[31] D. Whitley and A. M. Sutton, "Genetic algorithms-A survey of models and methods," in Handbook of natural computing: Springer Berlin Heidelberg, 2012, pp. 637-671.
[32] D. Thierens and D. Goldberg, "Convergence models of genetic algorithm selection schemes," in International conference on parallel problem solving from nature, 1994, pp. 119-129: Springer.