Printer Friendly

Quality-focused automation.

Automation has always been considered a productivity-improvement tool, with a side effect of improved quality. Although no one doubts its productivity advantage, automation is gaining increased-actually strategic-merit when the primary focus shifts to quality. The productivity benefits are still there, the only difference is that automated solutions are applied when quality issues demand them.

Focusing on quality means highlighting performance problems measured in quality terms. It means dividing problems into those that result from process incapability and those that result from process instability. According to Shewhart, instability must be resolved before incapability. Pareto has provided a diagramming technique that helps you isolate major incapability problems from minor ones. Deming blames the system itself, not the people, for violating the solutions to quality problems. Putting together the messages of these men of science leads us to conclude that there is a definite need for a strategy in identifying quality problems and finding automation solutions.

How are quality problems found?

The tool is Statistical Process Control (SPC). Used appropriately, it distinguishes between instability and incapability problems, and the Pareto chart helps quantify more important from less important. Together, they identify the most important quality problems. Powerful statistical methods-based on a structured approach called Statistical Problem Solving SPS)-are available to solve these problems.

I will not describe the SPS solution process here, but assume that such an investigation will reveal one of the following seven types of quality problems and corresponding automation solutions. These are based on my experience solving real-life quality problems. Their frequency differs from operation to operation, but the inherent idea behind each of them does not change:

1. Operator-handling variability. The quality problem is caused by human handling that does not repeat exactly each time (transferring parts from machine to machine, loading, unloading, etc). The automation solution is the installation of transfer lines or robots.

2. Operator-judgment inconsistency. The problem is that human attention and judgment do not repeat exactly-operator judgment is not consistent with process knowledge. When an acceptable range of quality is exceeded, needed process adjustments are not made. The automation solutions can include programmable controllers, automatic tool changers, or automatic part cleaning.

3. Material variation affects process response. Changes in incoming material properties affect the process and, ultimately, the resultant product (for example, harder or softer parts require different machining responses). The automation response is adaptive controls.

4. Material variation affects resultant product.

Changes in incoming material are directly responsible for unacceptable product variation. The automation response must be a higher order of adaptive controls; i.e., artificial-intelligence, backward-logic-compensating or "forgiving" controls.

5. Intermediate indications of product failure. An intermediate process dimension or out-of-range signal readily indicates the need for process correction to prevent making a bad finished part. Automation solution is a basic gaging system monitoring that condition with feedback controls to correct the process.

6. Higher level of process control required. Product property variations are more complex and neither plotted nor statistically interpreted. The result is process-control overreactions or inactions. The solution is automated SPC. For multistation machines, pinpointing the output of each station may be required.

7. Situations requiring 100% inspection. When safety issues are involved or a single unacceptable product cannot be tolerated, human inspection cannot be relied upon. (Or even the risk that sloppy housekeeping will result in rejected parts becoming mixed with accepted parts). The automation solution is totally automatic product-inspection and process controls.

Automating quality improvement

In recent years, a host of improvement ideas have deflected management attention (and their financial resources). Companies focusing on one quality-improvement method try to fit every problem into that solution. Companies automate SPC without realizing that on-line SPC defines problems, but does not necessarily solve them.

One cycle of improvement that is very effective uses as its seed a manual SPC program. Starting with inexpensive manual methods to find out what first to do, you add to this a small initial investment in automating the solution-finding process (adding more advanced process measurement methods). You then automate the indicated solutions to lock in these quality gains, and use these gains to finance further improvement.

Automation is not necessarily the best return on investment if it is not applied with strategic guidance. The initial manual SPC-by defining the problem-provides that guidance. The problem-solving process reveals which type of automation is needed, and that solution is applied. Strategically guided by SPC, each application brings in financial benefits, both in quality and productivity. Reinvesting the gains in automating SPC itself speeds up the problem-solving process. The end result is automating problem solving itself.
COPYRIGHT 1991 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Bajaria, Hans
Publication:Tooling & Production
Date:Mar 1, 1991
Previous Article:Workholding debate: roller cam vs toggle.
Next Article:Job two.

Related Articles
Plug and play to keep down costs.
Army publishes 2007 posture statement.
AMT establishes office in India.
Enhancing the laboratorian's role through automation.
Variety is the spice of life--for niche market sectors: six trade shows in one venue may be a lot to take in, but Mark Temple-Smith, Events Director...
Modeling, control, simulation, and diagnosis of complex industrial and energy systems.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters