Safety and Quality: 'Must Have' Metrics for Continuous Improvement
In my last article, and the first in our series on metrics, I made three key points:
- Embrace the notion that most factories are cost centers.
- Create the mindset of CI/lean, i.e., there is always a better way. Our job as leaders is to help our people find it, involve our people and plan the attack on the highest impact issues first, and provide training on the tools necessary to succeed.
- Assess your capability with soft infrastructure to support your need for actionable data. This is sometimes easy and at other times has long lead times. Get data collection and analysis moving early.
In this and other upcoming articles I'll suggest what my "must have" metrics are based on 40-plus years of experience in manufacturing. The first one is Safety.
Safety is often the easiest thing to measure because OSHA has provided the base formulae for key metrics over a long period of time. There are three specific measurements to track and use to drive improvements.
- Recordable Incident Rate: Number of Accidents in a specified time period x 200,000 ÷ Total Number of Hours Worked for the same time period.
- Days Away/Restricted or Job Transfer Rate: (Also known as DART and a relatively recent add) Total Number of Incidents that had one or more Lost Days, one or more Restricted Days or that resulted in an employee having to transfer to a different job within the company due to the injury. Multiply this number by 200,000 and divide by the number of employee labor hours at the site.
- Cost per Recordable Incident: Some measure the cost by taking the Workers' Comp cost and dividing by Hours Worked to get a cost per hour. I prefer to track Cost per Recordable Incident, which indicates the severity of the injuries. This is a more compelling number when communicating about the both the personal and business impact of having accidents. It's simply Workers' Compensation Costs divided by the number of Recordable Incidents for the month and YTD.
Safety is the one topic everyone can agree on, so it can be a positive driver of behavior and culture change. Whether the employee is union or non-union, hourly or salaried, regular or temporary, it doesn't matter. Nobody wants anyone getting hurt. Of course this starts with the communication of expectations from the leaders as well as their own behaviors matching up with the stated expectations.
Many of the plants I visit like to show how good their safety record is when compared to the industry average. Please stop this nonsense."
For example, anyone in a leadership position must never walk past a safety hazard, must always pay close attention to detail during Gemba walks, and must deal with safety issues with demonstrated urgency. In short, the leader must be the behavior model others are expected to emulate. All leaders live in a fish bowl 24/7/365. Everyone on the shop floor is watching you. Your leadership and metrics on safety will help you start a grass fire that "things are changing around here"--and the culture change you seek will begin in earnest.
One final note on safety: Many of the plants I visit like to show how good their safety record is when compared to the industry average. Please stop this nonsense. Unless your industry score keepers will identify for you the companies that have the lowest incident rates so that you can contact them to benchmark, then this information is just a distraction. To say your Recordable Incident Rate (RIR) is better than the industry average is irrelevant. For example, let's say the industry average RIR is 4.0 and your plant rate is 3.2. So what? World Class safety numbers have progressed over the last 10 years or so from an RIR of ≤ 1 to what some now claim to be ≤ 0.5. There are likely numbers in that neighborhood by the very best in your industry. Always keep in mind the big picture: The objective is to have zero recordables first and then zero incidents, period.
Let's Talk Quality Failures
The second set of metrics I'll tee up is quality failures. Most companies I visit do a fine job of reporting the costs associated with scrap, rework, customer complaints, returns and allowances/adjustments. However, most of them don't have reports that can easily be sliced and diced to expose the issues where focus should be first. Specifically, knowing we had $500,000 worth of scrap last month is important information in terms of understanding the size of the opportunity to improve the business. But it's looking in the rear view mirror and isn't actionable with any specificity. What does a first line supervisor or process engineer do with this information?
It begs the question again regarding the homework assignment from last time. What soft infrastructure do you have, or that you need to go get, in order to move as close as you can to real-time reporting -- sorting via Pareto analysis, by machine, by shift, by defect code, et cetera? Some of you may have electronic data collection and analysis capability from state-of-the-art equipment that have been programmed based on your needs. Unfortunately, most do not. If you can develop the capability short-term, great. If not what will you do instead? And most importantly, what will you do about this NOW? Really.
Do you have other options through your MES or CMMS systems? What can you do manually to get hourly data from your operators? What can you collect in batch mode so you're dealing with hourly information or, worst case, daily?
Your homework for next time:
1. Finish assessing your facility's capability to collect data at a level of detail that allows prioritizing the causes of quality defects and to initiate corrective actions. Again, what will you do NOW? Do not allow the collection and analysis of the data to be the constraint to the improvement process!
2. Find out how you are calculating First Pass Yield. This is a critical metric that must tell you the truth about how your processes flow.
P.S. Coincidentally, the latest issue of Target magazine (a publication of the Association for Manufacturing Excellence, AME) has an article penned by the CEO of The MPI Group, John Brandt, that you should read. It's called Knowing is "step one" to improving. To share a couple of quotes from the article: "The first step in fixing a problem is in knowing that it exists. The faster your staff can identify a problem--ideally the instant that it happens--the faster they can solve it......Unfortunately, most manufacturers can't identify problems as they happen even for basic objectives such as quality or safety."
A bad system will beat a good person every time. -- W. Edwards Deming
A relentless barrage of "why’s" is the best way to prepare your mind to pierce the clouded veil of thinking caused by the status quo. Use it often. -- Shigeo Shingo
An environment where people have to think brings with it wisdom, and this wisdom brings with it kaizen [continuous improvement]. -- Teruyuki Minoura
