The 9 Biggest Challenges to Scheduling Your Machine Shop and Why Most Schedules Are Dead on Arrival!

I recently surveyed 1,500 NTMA (National Tooling and Machining Association) machine shop owners about the biggest challenges they face when trying to schedule their shop. And then I spent countless hours going through the data.

I learned a lot and I really felt your pain. But by sharing my findings with you, my hope, is that you will learn a lot as well and maybe we can ease some of that pain.

The 9 Biggest Challenges …

Based on my survey (and from many of my clients and those I meet while Dr. Eliyahu Goldratt’s Global Marketing Director) I can say this… you are not alone. With all the responses I received, I could categorize ALL of them into 9 challenges. So let’s go through each category …

1. Clients change their mind

Customers often want to make changes after they have placed their order. They want to change their quantity, change the scope, change the design, cancel, or ask you to give priority to another one of the jobs.

In addition, customers often have emergency needs. And big or important customers always seem to pick the time when you’re booked solid for 3 weeks to call and request that you slip their job in this week so that they can meet a commitment to their customer.

So, we break a setup, we jump through hoops and do what we need to do to keep the customer happy. But we are now late on another customer’s job. And now the schedule is out of date.

And sometimes, they don’t want to make a change, but they want to check in with you to make sure your going to finish their job on-time.

And, we hate answering the phone.

2. Vendors are not always reliable

Raw material suppliers, particularly for less common materials, can extend or vary their lead-times and then still don’t always deliver when promised.

In addition, outside process vendors like platers, heat-treaters, welders, and the like also have a hard time meeting their original commitment. And the amount of time they end up taking varies based on the load on their facility so you can’t even begin to predict.

So as soon as one of our vendors misses their commitment, our schedule is out of date, and we now may be late. And when it finally shows up, we find ourselves breaking a setup and expediting because THAT customer just called.

And, these guys seem to schedule based on who’s screaming the loudest, so we need to call and scream on a regular basis for any important jobs. This isn’t very productive work, but it’s necessary.

3. Our mix can vary wildly and so our constraint moves

It seems that the nature of a machine shop is that your mix is going to change, and due to that your constraint (your Herbie) is going to move. And, it’s hard to improve a moving target and improving everywhere is just plain unaffordable.

Machine shops can range in the amount of repeat work they do. And the more custom work they do, the more the mix can change day-to-day or week-to-week. And sometimes “the mix gods” are good to us, but sometimes they’re not and we don’t ship on-time.

But shops that do a large amount of repeat or make to stock work don’t have it any easier — they are trying to balance against an unknown and ever changing forecast. And when the forecast is wrong, we need to break setups and expedite.

4. My employees do not always have the right skill and their discipline is lacking

Machine shops universally seem to lack skills amongst their employees. Certain people have to run certain jobs. Not everyone can do a setup, or not everyone can do certain setups. I’ve heard some owners refer to their less skilled labor as button pushers or part changers.

This means that the skilled people always have a backlog of work while we chase around and find something the button pushers can’t screw up. And finding the time to cross train is difficult.

And in addition to these skill issues, people don’t always show up on time or at all. Or they show up, but you wish their attitude would have stayed home…getting buy-in is difficult at best.

5. My processes are not reliable

Even jobs that repeat can have large differences in set up and run times depending on who is running it or what machine it runs on. But even if we can get the same guy and the same machine each time, stuff happens — tools break, fixtures don’t work, machines or tools aren’t calibrated, etc.

And then we win a new job, one we have not run before. And, of course, it doesn’t run anything like what we planned or based our pricing on. Stuff happens.

Our front office or pre-manufacturing processes are not much better. This means that some of the time we don’t always have what we need to run the job — the raw material, the design, the order into the system, the programming done, etc.

And when stuff happens (or more correctly when variability happens) we are in jeopardy of delivering late, which causes us to expedite, which means our schedule is out of date.

6. Machines break down

Of course, on occasion, machines break down. It’s difficult to schedule maintenance when you’re always behind schedule. And it seems to happen when we’re slammed. And then our schedule is out of date, we are in jeopardy of delivering later, and so we again expedite. (We’re getting pretty good at this fire fighting and expediting.)

7. Quality is not near perfect

Quality isn’t always perfect. Sometimes we struggle to make it right the first time or don’t catch a mistake until further processing has been done. Regardless of whether we spend time doing rework due to an unpredictable downstream activity or an unsuccessful tool tryout the result is we are in jeopardy of missing our due date, which causes us to expedite, and now the schedule is out of date.

8. Our data is not readily available nor accurate nor communicated

It is difficult to predict the load on our facility relative to our capacity — our reports and existing software don’t help or they are more trouble than they are worth. Our estimates for set-up and job run times are not accurate. This combined with all the items above make it difficult to provide due dates that we can hit 99+% of the time.

And because so many things can go wrong at any time, and we don’t have good feedback data or communication within the shop — we can’t predict. We can’t predict when we are going to complete a job, if we’re going to be late, so we end up breaking set-ups and expediting when we really need to.

And if we try to improve our due-date performance (DDP) by extending lead-times, we start to lose work to the competition. And if we miss too many due-dates or by too much, we are in danger of losing the client anyway.

So we do what we can. We use the reports we have which are based on less than perfect data. We create a detailed shop schedule, and then we update it, and update it, and update it …

9. Communication between silos is difficult

When something goes wrong within our company or with one of our vendors we don’t always know right way. Real-time feedback is non-existent. And customers don’t always get back to us in a timely fashion.

We don’t always communicate with our sales people and they don’t always communicate with operations/scheduling. There is usually just a lot of finger pointing back and forth. And a number of people want the ability to change shop priorities.

It’s not that we don’t want to communicate, it’s just that everyone is so busy dealing with all of the stuff above, there’s no time to do yet one more thing. And, who needs the conflict that is likely to occur?

We don’t have a quick snapshot of what we should be focusing on at any particular time. We don’t know what’s in jeopardy of being late, and it’s tough to get a sense of how we’re doing.

The Schedule is DOA …

So now we know why our schedule is Dead On Arrival. Actually when you list out all these challenges it’s really amazing that we do as well as we do.

So it’s no wonder it’s so hard for you to maximize your productivity, achieve 99+% on-time performance and reduce lead-times. But nevertheless, we do try to improve. The problem is that we focus on improving one or a couple of the above challenges and we diminish them some — but we don’t have any substantial impact on our on-time delivery performance or reducing our lead-times.

What if we had a Paradise Plant?

If we have correctly identified all the major causes for the difficulty in managing production, then it means that if we could address each one that the shop would be relatively easy to manage. The schedule would not change, we would not need to expedite, and we could be on-time, all the time.

Do you agree that, if:

1. clients never change their mind, 2. and vendors always supply whatever we ask for, on time, 3. and our mix stayed constant and our constraint did not move, 4. and people are excellently trained and disciplined, 5. and all processes are reliable, 6. and machines never break down, 7. and quality is superb, 8. and data is readily available, accurate, and communicated, 9. and communication is good,

Then:

ยท Managing production would be a piece of cake?

This seems logical. So I tried it. Now, I could not find a real Paradise Plant, but I did find a simulated one. And I gave the simulator to some very good schedulers. The simulator didn’t think for them, it didn’t make any decisions, it only executed the decisions they made.

The simulator presents a relatively simple operation — considerably fewer resources and products than what you have to manage in your real operation. And there is NO variability or skill issues. All of the challenges listed above are gone.

We went over how the simulator works – how to order material, how to set up a machine, what the orders are, the exact routing’s for each product, exactly how long each process is, etc.

The simulator could be frozen and the participants had as much time as they wanted to carefully plan and execute. They also did a short trial to make sure that they weren’t hindered by the software. This should be a piece of cake right?

What Happened in Paradise?

We asked each participate how they did and what their results were. NONE of the participants shipped all their orders — meaning they were NOT 100% on-time. And it wasn’t because of lack of capacity.

In addition, they reported that their well planned, detailed schedule was only good for about the first 2 days. After that they were running by the seat of their pants making a lot of course corrections.

How can it be that when all the challenges are removed, we are still left with the same 3 undesirable effects /problems?

1. Not all customers’ orders are shipped on time.

2. Original plans have a very limited life or are dead on arrival.

3. There are a lot of course corrections and expediting

This means that the list of challenges we collected are not THE major cause of the above three items because they occurred even after we removed these challenges. So it must be that we have not yet identified the true cause for these negative effects.

SUMMARY

So a better ERP, customers who don’t change their mind, higher skilled labor, a constant constraint, a constant mix, and so on wouldn’t completely solve the 3 problems.

But working on one or a few of these challenges is typically what we do. We hire a Lean consultant to help reduce our set-ups or to better organize our work space — and we do reduce our set-up time and we are better organized, but did we substantially improve our DDP, reduce the number of times we need to break setups or reduce our lead-times? The answer is typically no, not substantially.