Author: Paul Van Metre
Lots of shops have difficulty getting metrics on the quality of their planning or manufacturing engineering and then translating that into more profitable jobs in the future. It is a critical step to ensure that work flowing onto the shop floor is doing so as consistently as possible, therefore setting up the shop for success in executing the work. So what are some good key performance indicators? They say if you don’t measure something it won’t improve. I like to start way at the top to answer this. What is the ultimate goal of your company? It could be a number of things, but making a profit has to be part of the equation or else you won’t be doing it very long. In order to make a profit, you have to have more jobs that make money than lose money (at least in total dollars).
Net profit has to be your first metric. Few shops have good ways to measure net job profit in real-time, let alone after the job has shipped. In somewhat simple terms, there are two primary components to making a profit on a job. Keeping your dollars spent out-of-pocket to be less than you estimated, and keeping your DLH (direct labor hours) below the estimating target as well. For example, you estimate with a 10% profit target, we all know it doesn’t take much for your $ and DLH to exceed that 10% figure. So in order to have job profit be a metric, you have to measure it and then actually do something about it. It MUST be a closed-loop system. If you are over on a job, find out why (you will need good time tracking software that can categorize multiple categories) and learn from it and put something in place to reduce the chance of it happening again.
If you beat your targets it just as important to analyze why and try to replicate that in the future. With enough data on past jobs you can start seeing trends that may be related to part geometry (difficulty), tolerances (tight tolerances take longer), materials (it’s hard to be good at all of them), quantity (long set-ups will kill a job faster than anything, especially with smaller jobs), customers (some won’t pay as much), and industries (same issue as with customers). Analyzing these trends will help you make decisions at the highest level of the organization that can steer the company in a more profitable and successful direction.
Documentation quality could be your second metric. When it comes to preparing travelers/work instructions, there are two most fundamental rules. Eliminate tribal knowledge, and standardizing work. You need a system for disseminating all the information that was learned along the early stages of the value stream to the workers doing the machine set-up and running. That will come from salespeople, customer service, estimators, planners, purchasing, scheduling, QA, and possibly others depending on your company. You need good systems for capturing this knowledge of the job and getting it to the shop and it has to be able to evolve over time. We call this Part Development as parts are anything but static and the flow of communication definitely develops over time. Just as important is allowing your shop floor workers also put their knowledge back into the system. Allow them to document the set-ups with pictures, video, and text. Enable them to record challenges or problems they faced so it can be seen by everyone in the organization that can affect change in the future and they won’t be the only person that knows how to set up that job. After a while, you will have a rich knowledge base for helping troubleshoot future problems.
When it comes to standardizing work, the keys are to develop a robust system for doing things the same way every time and presenting all the needed information in the same format every time. So the metric here would be documentation quality. Allow the shop to issue an NCR or a CAR to the office – not as a spiteful action, but as a valuable tool to really have a two-way conversation about what is needed for the shop to successfully execute. Allowing this was a big revelation in our shop years ago.
There could certainly be many more metrics, but those two will get you going in the right direction. Best to focus on a few core ones than too many which muddy the picture.