Big Data Analytics in Supply Chain Management

4. Big data in logistics

Since the 1980s, optimizing resource consumption and outsourcing non-specialized activities such as logistics processes had been a major practice in most businesses. Today, logistics is a critical part of both the manufacturing and service industries. Many multinational companies have outsourced their logistics processes to third-party logistics providers and consider them their strategic partners.

Many research projects have been developed in the transportation and logistics industries that use available data that generated by road sensors, GPS devices and customers' websites. Logistics providers manage a high volume of product flow and have access to a considerable volume of data. Any measurable criterion of product flow - such as origin, destination, size, weight, price, load content, etc. - can be a valuable method by which to use information for value creation. Improvements in GPS efficiency, applications of sensor networks, and developments in the Internet of Things have opened up new areas in logistics and supply chain automation.

Optimizing service experiences such as delivery time, resource application, and geographical coverage are continuous challenges for logistics systems. Both delayed and early deliveries would be costly for logistics providers. This time difference between the planned delivery and the actual delivery is one of the key risk factors for logistics companies. Weather forecasts and vehicles' performance reliability data can be used to minimize the risk of inaccurate delivery times.

Big data logistics can be defined as modeling and analyzing logistics systems using big data sets which have been generated by GPS devices, cell phones, and the logistics companies' operations. Considering the current trend of big data applications in the logistics industry, it can be safely said that the logistics industry is in a transition phase from product-based services to information-based services.

There are important business processes in the logistics industry such as forecasting, transportation, inventory management, and human resource planning and management that can be improved by using big data. Jin & Kim combined big data analytics and business intelligence to minimize the analysis cost for the sorting and logistics processes of a courier firm. The other possible applications are forecasting delivery times, managing customer relationships, developing real-time scheduling, and managing supplier relationships.

Data mining applications for logistics management are threefold: 1- Data analysis to match the needs of both logistics process management and the customer; 2- Data analysis to manage the logistics process based on methodical decisions; and 3- A supporting role for logistics managers. Network technology developments can improve logistics processes by using information regarding real-time transactions. As a result of the increased adaptability of logistics information, logistics has been transformed into a dynamic data process. Niu et al. studied two competing air cargo carriers and showed how using big data analytics benefits a carrier by allowing them to receive updated demand signals. There are many quality papers in this field; some of these journal articles are summarized in Table 3.

Table 3 High-quality articles using big data in logistics processes.