Automobile Manufacturing Logistic Service Management and Decision Support Using Classification and Clustering Methodologies

Abstract

Given the growing complexity of consumer preferences and the underlying market advantages of addressing these preferences, manufacturers and logistic service providers constantly monitor supply chain efficiency and quality requirements. Third-party logistic services are offered as a means to attract customers and enhance competitiveness as long as these services are effectively integrated into the order fullfilment processes. This research uses customer preference attributes to define distinctive dilivery and distribution of orders. The clustering and classification methods provide decision support capabilities to logistics providers so that they can adapt processes to satisfy specific customer preferences. A K-means clustering algorithm clusters customers’ orders using demand attributes. Second, a decision tree classification approach analyzes each cluster segment using the history of consumer order preferences. Thus, the cluster results are the input data for the classification of logistics operations. The logistics service provider’s delivery services are tailored to satisfy each customer’s order requirements and preferences.