A Process Of On-Going Improvement (POOGI) - Part 7

Let’s take a simple environment as an example. Say that there are seven operations in this environment:

1. customer service receives and enters the order,
2. engineering programs the part, and when released the order processes through
3. lathes,
4. mills,
5. outside heat treat,
6. inspection, and
7. packaging/shipping.

Let’s assume the starting condition is a quoted lead-time of 4 weeks with due-date performance of 80%. Work-in-process (WIP) inventory is $100,000.

After an initial Goldratt Theory of Constraints analysis, we determine that milling is the constraint resource. In order to protect the mills from running out of work, we decide we always want a buffer of ready-to-process material in front of milling. And we want to release new work to the floor only at the rate milling is processing.

Since there is more WIP on the floor than desired, our first action is to not allow any new work to be released to the floor! We have a rule of thumb to reduce WIP to about half the level of WIP we started with. The focus is not starting new jobs, but finishing the jobs that have been started.

...to be continued.

Here's to maximizing YOUR profits!

Dr Lisa Lang

(c)Copyright 2008, Dr Lisa, Inc. All rights reserved.

A Process Of On-Going Improvement (POOGI) - Part 6

Of course, doubling available capacity must mean that we aren’t managing our current capacity as well as we would like, despite our best efforts to do so. Is this a reasonable possibility? Let’s check some “facts”:

Does your mix vary widely from week-to-week, as is typical of most job shops?
Do most jobs spend more time waiting than being worked on?
Is the touch-time for a job less than 10% of the quoted lead-time?
Are your lead-times relatively long, for example, are they measured in weeks instead of days?
Is your due-date performance less than 99%, as measured by your original promise date?

If you answered “yes” to some of the above questions, then there is a good possibility you can reveal substantial capacity with Goldratt's Theory of Constraints Drum-Buffer-Rope (DBR) scheduling.

But first let’s see if your like most traditionally managed job shops. If you are, you have excess work-in-process. In an effort to keep resources—machines and people—busy, jobs are released to the floor as soon as possible. Queues of work are in front of many machines. Once a worker has finished the job they were working on, the next job is available to get started on. This is done in the name of high efficiencies.

What are the problems associated with this traditional approach?
First, having too many orders on the shop floor masks priorities, promotes local optima behavior and therefore prolongs lead-time and significantly disrupts due-date performance. Second, hectic priorities (hot, red-hot, and do-it-NOW) cause chaos on the floor; not having a priority system can cause some orders to still be late. Finally, in many plants, there are bottlenecks that prevent attainment of 99% due-date performance.

...to be continued.

Here's to maximizing YOUR profits!

Dr Lisa Lang

(c)Copyright 2008, Dr Lisa, Inc. All rights reserved.