An Optimization Approach Based on Machine Learning Algorithms for Shipping Lead Time Prediction: A Case Study

Many manufacturing and service companies face long shipping lead time due to disruptions and delay of products delivery, which may lead to customers being unsatisfactory. Due to delays and disruptions in products delivery, machine learning algorithms play an essential role in order to predict shipping lead time by implementing an accurate predictive model to overcome difficulties associated with delays in products delivery. In this project, we applied machine learning algorithms including logistic regression, decision tree, random forest, and support vector machine to design and build an accurate predictive model for food supply network to predict shipping lead time. The algorithms are compared to get the best accurate model based on the logistic data used in this project. Two performance matrices including mean squared error (MSE) and root mean squared error (RMSE) are used for the comparison between algorithms to get the best accurate model with the lowest values of MSE and RMSE. The results show that the best accurate model can be obtained by using logistic algorithm with the lowest MSE and RMSE of 4.27 and 2.06 respectively. Figures also show the model loss and model accuracy for each algorithm used in this paper.