Reshaping the Production Line with Big Data Analytics

Dark data – this refers to the data collected from the shop floor that remains untouched or used to improve manufacturing output in any way. The average enterprise’s leaves 90% of its data sets untouched and this applies to facilities within the manufacturing industry. This statistic means that despite the increased rate at which data collection solutions are being deployed, most of the captured data sets remain in unused silos. But this shouldn’t be the case.

 - Success in Simulation and SchedulingThe paradigm shift to the smart factory requires automating traditional workflows and putting dark data to work using cutting edge technology solutions. Manufacturers who envision a future where interconnected machines interact in real-time and take accurate actions must understand that dark data will eventually have its uses with consistent advancement in technology.

Two decades ago, analyzing machine data patterns to predict failure was barely possible due to limitations with collecting specific data sets such as vibration data. Today, the sensors and IoT devices required to collect machine data from even legacy equipment cost a few dollars and it’s globally available. Conversely, gaining insight from dark data using technologies such as digital twin technologies, APIs or IIoT platforms will become the norm in the coming years and manufacturing enterprises must position themselves to take advantage of expanded big data analysis capabilities.

Accelerated advancements in big data analysis technologies will provide manufacturers with the means to develop data-driven strategies to reshape the shop floor. Use cases include:

Capacity Planning and Layout Optimization

Data from the factory floor always has a story tell. The story could be one of improved machine performance or the effects of increased demand. Leveraging the correct analytical tool develops the outline of the story, its plot, and finishes it. With accurate data and the proper analytical tool, manufacturers can anticipate change and develop optimized plans to deal with expected changes.

Capacity planning provides manufacturers with an opportunity to leverage data from the shop floor. Here, a capacity planning tool uses historical data to determine how fluctuating demand will affect available resources and how to mitigate operational challenges. Capacity planning also ties into layout optimization because shop floor arrangement reduces shop floor traffic and improves operational performance.

Traditional capacity planning tools such as manufacturing enterprise systems or simulation modeling software are equipped to improve capacity plans. Utilizing the digital twin brings a real-time element to capacity planning as CKE Restaurant’s example shows.

To improve productivity CKE Restaurant, the parent company of Hardees and Carl’s Jr., planned to reconfigure its kitchens layout and improve their capacity to meet increased customer demand. Leveraging shop floor data, CKE developed different configurations of proposed shop floor plans with the capacity to improve productivity. Employees successfully interacted with the digital environment using augmented reality headsets to test and determine the optimal configuration for its proposed store layouts.

Condition Monitoring and Predictive Maintenance

The most common application of the data collected from the shop floor is to keep track of machine and production processes by continually capturing operational data. Condition monitoring empowers the manufacturer to detect fault lines and requires extensive analysis. First, data collection technologies such as IoT devices or sensors keep track of machine operations and analytical tools constantly analyze data patterns to detect anomalies. Generally, detected anomalies are then pin-pointed or traced to the erring machine component and action is taken.

Condition monitoring is also the precursor to developing predictive maintenance strategies. The data captured from condition monitoring deployments become the historical data sets that are analyzed to discover patterns that predict future equipment failure. An example of predictive maintenance is Volvo Group Truck’s deployment of an IIoT platform to track and analyze data from its trucks. Constantly monitoring truck performance rate and vehicle data helped Volvo develop a predictive maintenance strategy that reduced diagnostic times by 70%.

Supporting Industry 4.0 Applications

Implementing Industry 4.0 revolves around understanding its core business concept. These concepts include; data-driven plant performance optimization, predictive maintenance, validation and testing etc. Successfully adopting any of these business models requires extensive data capturing and analytical capabilities. CKE Holding’s example highlights the use of data analytical tools to validate, test, and improve plans and Volvo’s example showcases data analysis as a tool for condition monitoring.

The industries move to Industry 4.0 is powered by digital transformation solutions that aid data capturing and data analysis but more can be done. Today, 65% of dark data remains hidden within manufacturing equipment and processes. Successfully harvesting or capturing these data sets will provide manufacturers with more fodder to work with to achieve industry-specific goals.


The introduction of edge computing is expected to ease challenges with capturing dark data from the shop floor and putting it to use. Simultaneously taking advantage of the decentralized analysis the edge provides and utilizing centralized analytical platforms such as the digital twin will advance the push to achieving the ‘lights out’ factory of the future.


Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.