Manufacturing Systems

A manufacturing plant with over 100 different machines and more than 150 operators, producing over 1200 distinct products. The customer was experiencing a large number of orders not being competed on time, without a clear cause. AnalyzeData used a variety of it’s techniques to identify and eliminate the root cause of these delay. Secondly, the AnalyzeData monitoring dashboard continues to be used to gain real-time insights into production efficiency.

Optimizing a complex manufacturing system of over 100 machines to increase throughput by over 20%
20% throughput increase

Manufacturing Systems

Process Mining

As a first stage in this business case , process mining was used to reconstruct the manufacturing process, and gain insights into the operations of the plant. Already in this stage, various anomalies in the plant operations were discovered (e.g. deviations from recommended machine types)

Anomaly Detection

Our state-of-the-art anomaly detection software made short work of finding the anomalous factors in the production process. Some of our key findings include:

  • Under-capacity of particular machines, allowing targeted upgrades to improve throughput;
  • Understaffed qualified personnel for certain machines, causing a bottleneck in the manufacturing process;
  • Over capacity, allowed the customer to reduce staffing in production stages, effectuating cost changes.

Results

Using our data analysis platform, the customer identified the root cause of their delays, and improved the throughput of their system by over 20%.

Continuous monitoring

Our real-time anomaly detection platform has built in monitoring dashboards. The customer continues to use these dashboards to monitor their process in real time. This allows feedback early in the manufacturing process, ensuring delays are dealt with as early as possible so delivery deadlines are not impacted.