Plant Productivity

Introduction

With an increased reliance on complex and expensive equipment, it is very important to maximize equipment productivity. This application provides a comprehensive overview of Overall Equipment Effectiveness (OEE) across plants and lines, and provides insights on how to minimize unscheduled downtime in manufacturing.

What is Overall Equipment Effectiveness?

Overall Equipment Effectiveness (OEE) is a high level measure of equipment productivity. This diagram shows equipment states, the types of losses that reduce equipment productivity and how they are grouped together in the OEE model.

The OEE model combines measures of equipment availability, performance and quality: 

• Availability is the percentage of time that the equipment is available to operate or Uptime. Scheduled downtime, unscheduled downtime and non-scheduled downtime (holidays or training) all contribute to availability losses.
• Performance is the speed at which the Work Center produces product as a percentage of its designed speed. Performance losses are categorized as either due to Rate or Operational inefficiencies. Rate losses are caused by equipment running slower than theoretical speed. Operational Losses are further broken down into Engineering and Standby Losses. Engineering losses occur when production turns equipment over to engineering, often to perform tests or experiments. Standby losses occur when the equipment is available but there is no product or operator to run it.
• Quality is Good Units produced as a percentage of the Total Units Started. This is sometimes referred to as First Pass Yield (FPY). Rework and Scrap also contribute to Quality losses.

Demo Walkthrough

Plant Productivity

In this demo, we have 3 plant locations across the world: Boston, Berlin, and Bangalore. On this page, you can select one or more of the cities by marking them on the map chart, and then view key KPIs for each location. These KPIs are the elements of OEE: Availability, Performance, and Quality, as well as OEE itself. With historical data, averages of these metrics can also be viewed over time on a monthly basis, allowing for additional time series related analyses and predictions to occur.

The plant(s) marked on this page will inform the insights generated throughout the rest of the demo. In this case, we have selected Bangalore.

Line Productivity

This page allows you to drilldown into each individual line at the plant that was marked on the previous page. In this case, the Bangalore plant has 20 lines; one or more of these lines can be marked to see the average availability, performance, quality, and OEE over time. After marking Line-ID 17, we see that availability is the lowest performing submetric, and it would be great to identify some root causes as to why this is the case.

Availability

On this page, we are focusing solely on the availability submetric, as we saw that availability at Bangalore-17 was not as high as the quality or performance submetrics, causing the OEE metric to drop too.

There are 3 types of availability losses, as shown in the donut chart below: scheduled downtime, non-scheduled downtime, and unscheduled downtime. Scheduled downtime refers to when downtime is planned well in advanced, and non-scheduled downtime refers to downtime due to holidays and training days. Unscheduled downtime is where the majority of issues arise - this is downtime due to unexpected errors. The goal here is to reduce the amount of unschedule downtime, by understanding the underlying causes.

When looking at unscheduled downtime losses over time, we can identify four main reasons these losses occur: pump failures, power interruptions, vacuum leaks, and changes in chemicals. This page of our application gives us some valuable insights with regards to these root causes; while there is variability in overall unscheduled downtime losses per month, the segments for power interruptions, vacuum leaks, and changes in chemicals remains relatively similar each month. However, we can clearly see that the variability over time comes from the amount of pump failures that occur each month. The months with the highest amount of unscheduled downtime losses (October 2022, February 2023, and June 2023), all see significant increases in the amount of pump failures. This would indicate to a technician that focusing on preventing pump failures and maintaining pumps more consistently over time will not only reduce the amount of availability losses, but will also make availability losses easier to predict in the future.

Compare OEE by Lines

On the last page of this application, you can now compare any of the four metrics we have been focusing on across multiple lines. We've identified root causes for availability losses for Bangalore-17, but using the filters panel on the right hand side, we can compare and drilldown further in any of the plants and lines that we have data for. Using the results of an ANOVA test, we can also gain a deeper understanding of our data relationships, and identify if any differences in availability, performance, quality, and/or OEE are statistically significant.