Shop Floor Optics: The Mystery of Production Leakage - Using Data to Find the Root Cause
Oct 08, 2018
By: Mark Stevens
Factory production of durable goods in the U.S. has grown in recent months and is 2.8% higher than it was a year ago. Seeming to contradict that trend is the factory utilization rate. This indicator of productive capacity stands at 75.9% and is 2.4 percentage points below its long-run average.
Not surprisingly, production managers and capacity planners are looking for ways to regain capacity utilization. Yet they’re perplexed as to where the operating variances occur, as well as how — despite adding more equipment and skilled operators, which is a growing challenge with the current labor shortage — profits keep slipping.
Let's take a look at a typical shop floor scenario, explore the possible causes and address how technology can speed up our problem-solving.
Painting the Scene: A Typical Plant Floor Scenario
Orders are coming in and volumes are up. The plant floor is busy with lots of activity, and workers feel good about how all the assets — people, facilities, equipment, etc. — are being utilized.
Behind the scenes, however, the VP of Operations is seeing a different picture. On-time delivery to customers is slipping, internal schedule attainment is off and work-center planning requires adjustments by the hour. Despite a bustling plant floor, the financials indicate that productivity is way off the mark, labor standards are being missed and pricing per unit is on an upward trend due to excessive labor variances.
Exploring the Possible Causes of Lower Productivity
Supervisors suggest that setup times are taking longer because of new, inexperienced operators. However, that conversation has been going on for months, and the new hires should be fully equipped to perform their job functions by now. Others speculate that the issue is with machine breakdowns, while yet another chimes in with concerns about materials handling and delivery.
After trying to address each potential explanation over the coming months, the results remain the same. Frustrations climb and conversations turn personal — all while the real cause remains elusive.
Uncovering the Hidden Clues
While the insights of operators and supervisors can certainly add value, there's no explicit data to support their suspicions. In order to uncover the hidden capacity, managers need the ability to analyze historical and real-time metrics to determine where and when leakages occur.
Achieving this type of transparency and visibility into production outcomes requires leveraging IoT devices such as Shop Floor Optics, a machine-monitoring platform designed to monitor production outcomes. A relatively inexpensive device is affixed to each machine to track and report historical performance, provide instant notifications of deviations via a dashboard or text alert and illuminate where true profitability and losses occur.
Real-Life Example of Shop Floor Optics Insights
How granular can the insights be based on the IoT data extracted from the Shop Floor Optics platform? Let’s take a look at some real-life dashboard screenshots below and dive into the various metrics to uncover the hidden capacity for our scenario.
The data indicates there are three work centers (CNC 104, CNC 108, and CNC 162) that comprise 23% of the jobs, yet account for 32% of the unplanned downtime. This represents a total of 2,762 wasted hours!
These insights created an “aha” moment for the team and helped them pinpoint which work centers to focus on first. With the data, operations could look deeper and discover additional insights to determine that:
- Every setup was ahead of plan, negating the assumption that setup times were the problem.
- The ideal parts per hour for production efficiency ranged from 42–66%.
- Machine performance was nearly optimal, ranging from 99–104%.
- The quality of parts was impressive at 100%.
Another dashboard clearly shows where the main issue lies:
The data clearly indicates that the main reason production is slipping is quite simple: over the course of 301 occurrences, no one was at the machine to produce the product when the day started. Addressing the initial explanation of lengthy setup times due to new operators could never remedy the situation because that wasn’t the problem.
Let the Data Tell the Story
With data in hand, personnel can make adjustments, ask the “5 whys” questions and assess why people aren’t at their work centers during production windows. Is the work center not staffed properly or is it being scheduled independent of available people? Are people on vacation? Is it a supervision issue? Once these issues are resolved, workers can move on to address other opportunities for improvement such as approval processes, machine downtime and setup times. And, of course, continuous analysis of the data can show when problem-solving measures are effective.
Production costs don't only mount when machines aren't running at peak capacity; major labor costs are incurred when multiple personnel waste time and effort making guesses and trying to address issues that aren't the root cause of the problem. Make sure to always ask questions with facts and explore with those closest to the work how you can improve and reduce production variation. Is it your methods, materials, machines, measurements or work environment? If causes to issues aren't immediately obvious, or their solutions aren't, don't be afraid to experiment and make changes. Circle back and look at the data to see if those changes are effective , learn and reapply.
Shop Floor Metrics a technology solution coupled with a productivity specialist that helps your organization see and identify leakage quickly and, more importantly, move to corrective action.Target the right problem by leveraging the right data. Reach out to Wipfli today for a free demonstration of Shop Floor Optics (powered by MachineMetrics) and to discuss we can help get you faster results and uncover capacity you didn’t even know you had.
“US Manufacturing Production Rose for Second Month in July,” Bloomberg, August 15, 2018, https://bit.ly/2IH3dAI, accessed September 27, 2018