The role of data in manufacturing has always been understated or unstated. The way companies cope with quality improvement has been transformed by new forms of data use and data analytics. The experts in the field report a considerable shift from exclusive dependence on post-manufacturing inspection work and retrospective analysis to the prediction and early identification of problem areas and maintenance requirements. New sources of data—from sensors to callcenter conversations—are bringing traditional product inspections on a new level. By transforming the management of quality and safety in asset-based businesses, these innovations are gradually improving manufacturing sector.
Data transforms technology, and it’s only the beginning of striking changes.
The quality and safety revolution in organizations was marked by numerous technical breakthroughs such as real-time data from connected vehicle sensors and GPS and text derived from warranty reports and conversions of callcenter speech conversations, just to name a few. On the other hand, the data is now combined in a repository that allows for multiple data formats and analysis across them.This is where exactly machine learning algorithms come to play. Their role is to identify trends in the data and to make predictions.
Why to use data mining?
Businesses use data mining to draw conclusions and solve specific problems. One of the key benefits of data mining is that it is fundamentally applicable to any process and helps improve the flexibility and efficiency of operations. Thus, data use in manufacturing facilitates schedule adherence, monitoring automation, modeling for capacity, and reduction of waste. The departments are completely transformed and factories become smarter by achieving full data transparency.
How manufacturing businesses take advantage of data mining
ABB, a huge manufacturer of a global importance, is currently using process mining for purchase-to-pay and production processes. Earlier, the employees from the ABB plant in Hanau, Germany, would extract evaluations from their SAP systems several times a day, import them into Excel, and use complex formulas to analyze and understand processes. Today, the relevant production and assembly team leaders at ABB receive an email in the morning that outlines the previous day’s production variants, throughput times, and number of rejections. As a result, the plant’s full ecosystem of quality improvement processes is immediately visible with process mining. The system only gets better at identifying patterns as more data gets fed in. Instead of relying on complex manual analysis of processes, operational processes provide instant results.
Drastic changes have impacted vehicle manufacturing industry too. In this sector, the products are relatively expensive, with high-end manufacturers focusing on service and product quality. They note that the business benefits related to the introduction of data-driven innovations have all the chances to speed up identification and resolution of quality problems, as well as cut warranty spending, which amounts to between 2-6 % of total sales in the automobile industry. For the customers and users of these vehicles and machines, early identification and preventive maintenance often results in greater uptime. For instance, in one case involving an automotive company, 28,000 vehicles were saved from recall by the identification of a problem before vehicles hit the market.
Data mining tools can be very beneficial for discovering interesting and useful patterns in complicated manufacturing quality improvement processes. These patterns can be used to improve manufacturing quality. However, data accumulated in manufacturing plants have unique characteristics, such as unbalanced distribution of the target attribute, and a small training set relative to the number of input features. Anyways, business process improvement has to start somewhere. Using an approach that incorporates big data, analytics and business intelligence approach is simply the most reliable, proven way to make improvements that last. Once you know what to measure, track it, analyse it, and improve it, you’ll have the right foundations in place to enhance processes throughout your business. Time and product waste will be the things of the past.