Using Big Data Analysis and Machine Learning Methods to Improve Key Performance Indicators for a Warehouse Company

Supply chain management is believed to become one of the main beneficial spheres from further development of computational and network systems. Therefore, the perspectives of using new techniques and concepts like Big data and Machine learning in different fields of logistics and SCM is actual and discussable. This paper aims on the application of this methods into process of improving warehousing performance indicators. By the literature review it was revealed that business analytics with combination with Internet of Things solutions is the most logical way to put these technologies into practice in warehouse logistics field. But the machine learning methods are not usually discussed in the context of managerial warehouse problems. Many studies refer to this theme but not many of them provide the particular solution of the specific logistics problem. For these purposes, the task of increasing the time of the warehouse picking operations by reconfiguring warehouse zones based on the cluster analysis of inventory items turnover is solved. The data has been generated by triangular distribution, but the same data can be collected in a real situation. As a result of solving the problem, a management solution was developed for reconfiguring warehouse areas, which, according to the calculated indicators, will lead to an improvement in the KPI of the order picking time. The benefits of using clustering along with traditional methods of FMR-analysis are described as well as the limits of the approach. The practicing such method as an alternative to the traditional one can be beneficial especially while using the turnover data as a base for FMR-analysis. The solving algorithm can be adopted by further academic researchers and the logistics managers in order to get more insights of increasing the operational efficiency. Keywords: warehouse logistics, supply chain management, business analytics, machine learning, big data, key performance indicators (KPI)