Where There's Smoke There's Fire

Where There's Smoke, There's Fire...

Seven Signals For Systems Surgery

Rick Duris’ "Systems Integrator's Notebook" - August '97

I love to pick people’s brains. I really do. I have two Clients who are masters at re-engineering manufacturing operations top to bottom, stem to stern. One of them used to work for a Big Six consulting firm, and the other is an expert at demand flow manufacturing. They seem to have an almost ESP/intuitive ability to diagnose what’s wrong with a manufacturing plant and then to implement the appropriate processes to achieve the optimum results desired.

To me, it’s an uncanny ability, really. I’m in total awe when I get the chance to see them at work. It’s like they’re wearing special glasses, through which only they can perceive/interpret the critical elements related to the work at hand. What’s funny is that I know it’s not textbook-type stuff, because some of the things they do seem to violate the principles of standard manufacturing theory (just like Michael Jordan seems to violate the laws of gravity).

Lately, people have been asking me about the process that I go through when trying to implement a successful bar code data collection system. They say that they’re intrigued by the questions I ask, the subjects/points I tend to focus on, how I interpret the results derived and how I determine the optimum strategy for them. They really say they find the process very enjoyable as opposed to sitting in another stupid systems meeting being bored to tears.

Shucks, I'm flattered-really. That’s quite a compliment. But I actually never really gave the idea much thought until I was asked to train someone from a client company on how to perform the studies which I go through for my Clients.

Signs, signs, everywhere signs

So, lately I have been giving a little thought to the things that I tend to focus on when I try to improve the data collection or system processes of a company. What follows are my technical "indicators". These indicators are "symptoms" (not causes) which alert me that something is amiss. My belief is that when you walk into a plant, warehouse or facility, you’ll find certain things which will immediately indicate that something can be improved here. (The old adage: "Where there's smoke, there’s fire" comes to mind.)

But before I launch into this, a couple of caveats are in order. Some of these indicators may contradict themselves. Also, be aware that they are not meant to be taken out of context either (i.e. All paper is bad.) These are just the focal points which I have learned that indicate a process/system may be able to be redesigned to produce a more optimal result.

#1. Paperwork: Lack of paperwork; too much paperwork; handwritten paperwork; faxed paperwork; forms; labels (on the warehouse racks and on the product at its various stages); redundant paperwork; stacks of paperwork; notebooks; Rolodex cards; memos; lost/unusable paperwork. All/any of it may matter when improving a process by implementing new systems. I scrutinize all of it intensely.

#2. Real Smart People or Managers Who Actually Do The Work: A couple of months ago, I wrote about Roy, who was the best production scheduler I’d ever seen. Companies that tend to create positions/jobs where managers (or other real smart people) are doing the work rather than managing the work do an incredible injustice to themselves and their people without realizing it. Managers should be instituting systems whereby ordinary people can do extraordinary things on a consistent basis.

They hold themselves back because this sort of strategy creates a sense of "indisposableness" which does not allow the position/company to change, nor the person to grow, nor be replaced when it comes time for a promotion. The person also tries to solve the problems by "being there" rather than creating systems where the work can proceed without them. In essence, s/he has one of the worst jobs in the world without even being aware of it.

In your quest to systemize a process, you potentially want the process/procedure to be capable of being executed by someone who’s last job was at McDonald’s. I’m not kidding.

Why? Because McDonald’s hires kids who can’t even clean their own rooms, yet because of the processes and training which McDonald’s uses, it is able to take these people, and within a few hours turn them into a cooks, maintenance persons, customer service persons, etc. And if you think that’s easy, think again. I know, because I used to be one of those kids.

#3. Un-Streamlined Processes: These include Redundant processes, batch processes and processes which have too much going in (input) and/or too much going out (output), regardless of the nature of the process. (You just know that somewhere between the input and the output is a real mess.) Usually, processes which handle lots of exceptions to the normal process are ripe for redesign. (By the way, if you start seeing lots of exceptions, you’ve got a tiger by the tail.)

#4. Non-Real-Time Access to Information: When people rely on information which is out of date the moment it is displayed/printed, you may have some real opportunities for improvement there. The usual problem when optimizing this type of process is that the cost to get real-time data to the person is sometimes prohibitively high. Regardless, it is worth your time to investigate.

#5. Information is entered into the system well after the process has been executed: These include processes where the collection of information is separate from the entry of the data into a computer system. Usually this process can be optimized to reduce errors, data entry labor, paper costs, etc.

#6. A process with highly variable results: If on one day production (or accuracy/errors, quality, usage) is high and then the next day it’s low…something’s up. You may be asking, "Why is this indicative?" The answer is usually because it is terribly inefficient to support such a fluctuating strategy. Try thinking in terms of setup, break/tear downs, scheduling, maintenance, forecasting usage/resources, anticipating effects on quality, people’s stress levels, etc. The more consistent the process’ results, the better.

#7. Processes which generate errors: This one is kind of like "low hanging fruit"…very easy pickings. Most data collection systems are usually put in to reduce/eliminate errors. Implicitly, you get a productivity boost just because you can eliminate back-end error processing/reconciliation.

Coming into focus

There you have it. Are there more indicators? Probably. These are just the ones off the top of my head. They’re the first things I tend to focus on when I get the chance to go into a new facility.

I hope this helps you. (I know it helped the person I’m training.) If you have some technical indicators for process improvement which you’d like to share, please drop me a line or give me a call. See you next month…

Rick Duris is the President of Business Technology Group, Inc., a systems integrator based in Libertyville, IL specializing in plant floor systems such as warehouse management system, bar coding and radio frequency applications. He can be reached by contacting RainMakers, 847-251-3327 or email: jon@rainmkrs.com