Big Data Analytics in Supply Chain Management
Read this paper for an overview and examples of how big data is used in specific areas, such as supply chain management, risk management, and logistics of business in industry. One of the biggest issues for analysts with big data is knowing how to separate the valuable data from that which does not help answer their requirements.
Sometimes people describe intelligence as "connecting the dots", but it is rarely simple like a "paint-by-numbers" art project. The dots are not just lying around waiting to be connected. More appropriately, it has been described as filtering out the right radio signals from the fray in a huge city. You have to be carefully tuned to your requirements, which will be discussed at length in Unit 2 and again in Unit 8, as these are the guide stars that keep you on track to finding the right data to answer the questions you need to focus on.
2. Big data in manufacturing systems
Manufacturing can be defined as the hard segment of an economy which applies resources such as labor, machines, tools, and raw materials in order to produce physical products. The manufacturing industry contains a huge volume of data created by sensors, electronic devices, and digital machines in factories. Manufacturing is a traditional industry which can be highly affected by big data, since the approach for many companies has been changed to operate based on forecasts. Moreover, big data could simplify data visualizations and improve automation applications regarding production design and engineering.
Manufacturing plants collect data using different channels such as manufacturing processes, supply chain management systems, and tracking the products sold. Using big data can help to develop new products based on customer needs. Moreover, manufacturers have the opportunity to better plan out their supply chain with a more accurate demand forecast. Managers believe that using big data can help diagnose defective products, improve process quality, and better plan supply chains.
Manufacturing processes can't be firmly separated from either logistics processes or supply chain management activities. For example, many of the logistics processes in manufacturing plants are performed by tools with radio-frequency identification (RFID) tags, which allows real-time tracking of the products. Using data analysis on the shop floor enables the system to efficiently implement real-time manufacturing, planning, and scheduling, which is directly affected by both the material delivery time and the real-time information coming from the manufacturing processes. Moreover, analyzing the big data can level the material flow and help the plant manager to better plan space limitations regarding material flow and warehousing operations.
There are a lot of process, personnel, and departments data generated during a product's lifecycle. The nine stages of a product's lifecycle were introduced by Tao et al.: product concept, design, raw material purchase, manufacturing, transportation, sale, utilization, after-sale service, and recycle/disposal. In each stage, a lot of data is generated, and by collecting this data for all products, we can have a dataset with big data characteristics. Five areas of big data application in manufacturing are: 1- using data to forecast a complex process's output; 2- using data to capture that which is difficult to measure under regular conditions, 3- developing algorithms which can more accurately control the quality and safety of the final product; 4- using image metrology to reduce the amount of human supervision required; and finally, 5- obtaining the optimal time for doing predictive maintenance.
The continued growth of the Internet of Things has also influenced the amount of data available to manufacturing companies. It has been forecasted that by 2025, about 175 trillion gigabytes of data will be available, and the manufacturing industry will be the second-fastest-growing sector for data generation, after the healthcare industry. In spite of this huge volume of data - which is generated and kept by manufacturing plants - the number of studies on big data applications in the manufacturing industry is still considerably less than that in the service industries such as finance, information technology, and E-commerce. However, despite the lack of big data studies in the manufacturing industry, data mining has been used frequently in manufacturing decision making problems.
There are several different areas of manufacturing - including new product development, smart manufacturing, cloud-based manufacturing, process improvement, predictive manufacturing, and redistributed manufacturing - in which the application of big data analytics can improve system outputs. studied the major contributions of big data analytics in manufacturing systems by examining several case studies. In order to have a better overview of the recent applications of big data in manufacturing systems, the Thomson Reuters Web of Science was used to categorize the most frequent big data studies in manufacturing. Figure 4 shows the relative frequency of published studies on big data analytics as applied to manufacturing. Studies are categorized based on their research focus in Figure 4, with most categories explained afterwards. It is worth noting that review papers are not being discussed further, since we are studying research contributions in the literature and "review" categories as mentioned in Figure 4 are just there to give further insight. Moreover, the "Marketing" and "Flexibility" categories are not discussed further, since each makes up less than 5% of the total publications.
Figure 4 Publication frequencies with "big data" in title and "manufacturing" in topic.