How Difficult is Management of Order-Driven, High-Mix, Low-Volume Production?

Many job shops apparently face a difficult task in managing order-driven, high-variety production with the help of very limited resources (which may consist of multi-functional machines and multi-skilled workers, possibly with different working hours). I guess the difficulty arises due to several factors like:

(1) Orders received with unpredictable process requirements at unpredictable times
(2) Short lead times imposed by some powerful customers
(3) Need to determine rational due dates and right start times for new orders based on the existing workload and resource capacities
(4) Simultaneous processing of multiple, diverse orders for maximizing throughput while meeting due dates
(5) Order sequencing for reducing the number of setups
(6) Long procurement times and unexpected material delays
(7) Vendor operations
(8) Acceptance of rush orders
(9) Changes in priorities, due dates, quantities or process requirements of orders
(10) Interruptions in resource availability
(11) Rework due to quality issues.

What solutions would you recommend to job shops for managing such complex production?

Whats low volume? 1 or 1,000? could be either depending....but if you mean a job shop which is more service than production, the complexity might be why most job shops are small - the stakeholder who really cares and who has over all view is hands on to parts going out the door. trying optimize/codify the solution to a very complex set of inputs that completely changes daily is extremely difficult - that's what the human brain is for....but it only works if things are small enough one human can see all.

Its the backflip business, everyday you're asked to do a new one (my experience is more fab than machining, same issues though). The bigger the organization, the more impossible it is to turn on a dime. The only solutions I have to doing this in a big larger shop are 1) get into a product type business and 2) on the one of's, target large projects, stuff that takes months, fso its worth it to turn the organization at it. One guy can have an idea and make it, 20 guys need drawings, supervision, QA, shop managers, IT, process, HR policies, admin,book keeping etc etc etc..... the $500 job just is not worth it. I'm at 21 guys now (started 3 years ago) and that's been what I've done.

....or I could have just said "you live with it"

With out knowing the items you produce, what materials you typically use and the available machinery, a person cannot make any accurate suggestions. Also with out knowing what you have in place that also leaves a person in the dark. In my horror of job hopping(8 shops)working for the man for 15 years I have worked at places where what I would consider simple common sense would increase production quite a bit. Worked at a place that had a department that had a dozen identical CNC Lathes that had 12 tool positions. None of the tooling positions were standardized. On consecutive jobs the center drill would move from T1, T5, T7 to T10. The place had been running those machines for 10 years when I got there. Standardizing tool positions saved an average of an hour on a well tooled set-up. The current machine I have owned the longest has had the same front turn tool insert holder sitting in position 12 since 2006, and the insert can be changed with out removing the holder.

To a degree its exactly like dualkit says, standardize what you can then juggle the rest.

IMHO its all about work flow not shortest cycle times in this sorta work. Its a very different mindset to production and too a degree with job shopping the more volume you have flowing the more you can mix and match stuff to make best use of a given setup.

Right now repairs are a premium cost option, but its also were you can make any number profit wise you want.

Predicting the unknown's come with experience, hell i have a job here right now that should have been done 3 months back, but the simple approach i was hoping to work turned into a monster big problem, net result instead of a simple pressing operation i had to build something far more complex to do it, great part is thats worked really well and the end results got another customer having hot flushes about how he can sell that to his customer base (you know you have cracked it when the ofcuts of a job are selling for as much as the job :-)

IMHO job shop non constant production works akin to surfing, some times you have to swim, some times you take a pounding, but some times its all worth it.

Skilled help is a must though, you gotta be able to pass stuff to others to do.

"....1) get into a product type business ..."

Lol...The holy grail! Trying to do this now as my last great customer isn't anymore...

You'll also find some guys are wired for prototype and some for production....put the wrong horse on the wrong track and you get a melt down. Find what your guys are good at and keep in that environment.

ShopPLanner said:

Many job shops apparently face a difficult task in managing order-driven, high-variety production with the help of very limited resources (which may consist of multi-functional machines and multi-skilled workers, possibly with different working hours). I guess the difficulty arises due to several factors like:

(1) Orders received with unpredictable process requirements at unpredictable times
(2) Short lead times imposed by some powerful customers
(3) Need to determine rational due dates and right start times for new orders based on the existing workload and resource capacities
(4) Simultaneous processing of multiple, diverse orders for maximizing throughput while meeting due dates
(5) Order sequencing for reducing the number of setups
(6) Long procurement times and unexpected material delays
(7) Vendor operations
(8) Acceptance of rush orders
(9) Changes in priorities, due dates, quantities or process requirements of orders
(10) Interruptions in resource availability
(11) Rework due to quality issues.

What solutions would you recommend to job shops for managing such complex production?

Click to expand...

At risk of sounding like a crotchety old smartass, this sounds like either a plea for help in figuring out how to run a shop with no muss/no fuss (not gonna happen), or another run at the "what software features would you like to see in a product I'd like to sell you".

You have a list already of a number of the primary issues facing daily operations in a job shop. If you are looking for automated help, there are hundreds of software packages out there that may help in some of these areas. The only issue with employing them to maximum effectiveness is in the area of updating the information the computer gets to see. If the info is not up-to-date, your output solutions will be ineffective as planning and execution tools. Updating the info generally requires people dedicated to THAT task, not making parts, unless you can afford a full-blown production tracking system with barcode readers, automated download of part movement data, automated scheduling of orders received based on machine loading and projected completion times, blah blah blah...and then you still need people to maintain the system (likely at a much higher level).

Job shop operation in my experience is really a best-guess methodology at any given point in time; those guesses are informed by perception of what's important at that time. You can plan all you want, but the external factors are always wild cards in planning. In a job shop, there's a lot of them.

Thanks for all the responses so far. For improving production management in job shops, there are some popular approaches like lean manufacturing, six sigma, theory of constraints (TOC), quick response manufacturing, MRP, advanced planning & scheduling, etc. How helpful are these methods individually for dealing with the job shop issues described earlier?

ShopPLanner said:

Thanks for all the responses so far. For improving production management in job shops, there are some popular approaches like lean manufacturing, six sigma, theory of constraints (TOC), quick response manufacturing, MRP, advanced planning & scheduling, etc. How helpful are these methods individually for dealing with the job shop issues described earlier?

Click to expand...

This sure sounds like the "blah blah blah" that specfab is talking about.

Mcgyver said:

Whats low volume? 1 or 1,000? could be either depending....

Click to expand...

I cannot find a specific number to determine the volume of an order as low or high. However, in my opinion, a low-volume order moves through shop floor as a single batch while a high-volume one moves in terms of transfer batches. Some job shops have a mix of low-volume and high-volume orders.

At risk of sounding like a crotchety old smartass, this sounds like either a plea for help in figuring out how to run a shop with no muss/no fuss (not gonna happen), or another run at the "what software features would you like to see in a product I'd like to sell you".
SpecFab 2016

Hey ShopPlanner...Are you a software distributor? I just went and checked all your posts and most everyone of 'em are recommending software that seem to address the issues you ask about in this post. I only checked 'cause I don't want SpecFab to be so hard on hisself....lol

In this thread, I would like to know the relative contributions from various approaches to resolve those difficult issues in job shop production management. I want to see a very open discussion (say brainstorming) about them, if possible. I prefer to not propose any specific method in this thread.

ShopPLanner said:

In this thread, I would like to know the relative contributions from various approaches to resolve those difficult issues in job shop production management. I want to see a very open discussion (say brainstorming) about them, if possible. I prefer to not propose any specific method in this thread.

Click to expand...

You should run for office. That was a perfect display of dancing around the question.

Merry Christmas

smallshop said:

Lol...The holy grail! Trying to do this now as my last great customer isn't anymore...

Click to expand...

it is the holy grail, and i wasn't meaning to be glib, just that those were the only two ways I could come up with for growing past the small job shop where one brain deals with all the complexity. I've done it a few times (product launches) and the idea for the product is 1% of it. Everything that goes into execution including organizational readiness, systems, sales, marketing, etc is the 99%. It has to be thought through and dealt with.

Widget making is very different from the job shop business, even if it still uses the same people, real estate and equipment.....so end of day I'm saying like Doug, "you live with it" unless you either move toward products or very large one ofs with their project manager

Hello ShopPlanner,
By having identified (in detail)the factors that you face, you are already on the right track. Unfortunately, there is not a "push button" method of preparing for these factors. A well organized highly skilled shop staff that works as a team is the only way to succeed.
otrlt

I agree with adama, ...............that said being Brits and living in Britain, we can always come out with the classic line when customers get stroppy.

Try to convince the customer to "pull in" orders to make them
larger qty's at less frequent intervals.

Usually by giving qty discounts, that reflect the set-up time.

So what royalty do we get with your software product planning endeavor?

The last shop I worked at the new owner thought the $$$$$ investment to have these suit and tie guys who worked mainly with production
shops come in and lecture us how to run our job shop would be a great idea.
Set us up with quoting and scheduling software using Microsoft Office Pro or some shit like that.

Right out the gate, boss didn't like the quoting software, plug in all the details such as material, tooling, labor and then the software was supposed to generate the quote based off of our overhead to make a certain % profit.
Boss pulled the plug on that and we went back to how we have quoted for years, calculator, pencil and paper and knowing what the max the market would bare for that part.
Ran several known jobs through the quoting software and it quoted 10-30% less than what we sold the parts for.

Then came the scheduling bullshit.
Boss wanted to log on from anywhere and at a glance see where we were on job XYZ.
All that was seemed nice during the lectures, on the overhead view screen anyway.

Reality set in.
The software means nothing if you don't consistently keep it updated.
Most of you know all to well that when you have employees to deal with, production issues, ordering, bottle warming etc. you don't have time to spend setting there updating all these little tabs in the program and being a small shop (15 people) you cant afford to hire a person who manages just the work flow charts, now your paying another employee just so you can use your fancy software package.

Now to add to the problem. This shop was known for rapid turn around, you got the money we got the time, got a 12 hour job you need in the AM? No problem.
Scheduling I be damned!

IronReb said:

you don't have time to spend setting there updating all these little tabs in the program

Click to expand...

This has to be the biggest problem of ERP (or other blah blah blah acronyms) software. The interface is truly horrible. From the biggest enterprise players to the little guys, it takes forever to do anything. It can't all be on one page. Oh no, it must require 4 screens and 12 tabs on each screen to enter any useful information. It requires looking at 6 screens to get any useful information back out. Then every time you want to quickly update a piece of information, you need to sit down and type in your username, then type in your password (that needs to change every 3 weeks "for security"), then wait for the software to start, then navigate 4 pages to make one little update.

The essence of these systems is:

1) Input data
2) Store data
3) Display data in useful combinations

My experience is that they do a good job of 2, but a terrible job of 1 and 3.

If the software was any good at 1 and 3, it wouldn't require hiring an extra person to operate it compared to the Paper/Excel/Whatever system that was used prior to buying the ERP system.

Compare that to CAD software. As much as Solidworks has a number of shortcomings, there is no question that they have put a lot of effort into making the user interface more efficient. In the 10+ years I have used that software, they have added numerous features to allow the designers to get designs done faster. Reducing the number of menus that need to be opened, reducing the number of clicks to get a task done, even reducing the amount of mouse movements, including templates and data reuse that eliminates retyping data. I haven't seen anything remotely comparable in the ERP world.
</rant>

Well as far as tracking, I have seen the paper job trailer having barcodes for each operation.

Operator simply scans barcode when his job is done, then send the job to the next op.

So that is surely simpler, but probably takes more time in the office to set up
to work properly.