| || |
Deming\'s 3rd Point
"Cease reliance on mass inspection to achieve quality. Eliminate the need for inspection on a mass basis by building quality into the product in the first place."
“Cease reliance on mass inspection to achieve quality”: From software’s point of view: Try to reduce internal builds. It tells that ‘Before sending build to QC check it thrice’
Please explain. This is very hard to map with Software
Re: Deming\'s 3rd Point
<BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>Originally posted by AmitK:
Please explain. This is very hard to map with Software<HR></BLOCKQUOTE>
No, no - this is actually very easy to map with software. Remember that, above all, Deming was talking about quality information in a variety of contexts.
So when he says "cease reliance on mass inspection to achieve quality" what he is saying is that this is not the most quality-use of information. In other words, the information derived from inspections is not of the best quality given other means of gaining such information. Inspection, in and of itself, does not improve quality in the best fashion and it certainly does not guarantee quality because inspection is sometimes done too late in the process. When it is done too late, the quality, whether that be good or bad, is already in the product to some degree. Thus it can be reactive. Instead of using "reactive" inspection techniques, Deming was advocating more proactive techniques - just as we now advocate that QA gets into, say, the requirements phase rather than just coming in at the system test phase. Or saying that QA only "inspects" requirements and does not help with the crafting of some requirements. Or that developers only "inspect" their code, but not have standards in place to guide their coding in the first place. Also remember that Deming was speaking out against some bad practices of sampling that were used in Statistical Process Control (SPC) and yet advocating others at different stages that were more indicative of the "true quality" of the product.
The whole point of Point 3, from Deming's point of view, is that "true quality" comes from the improvement of the process(es) to eliminate defects rather than from inspection. In other words, design good quality in rather than inspecting bad quality out. This is also based on the fact that the main "costs of nonquality" (to use Philip Crosby's phrase) are from the "scrap" and rework of defective products of different phases.
Another point of Deming's (and one that was echoed in later years by Larry English) is that many inspections (of the reactive variety) are quite expensive as a means to quality assurance because they are expensive (in terms of time, effort) and they are ineffective because they cannot assure one hundred percent discovery of defects, and these do not achieve real process improvement. Now, bear in in mind that Deming was using "inspection" in a more strict sense than many people currently do but also keep in mind that the very word "inspection" does imply a finished product (even if only in stages) to inspect. Thus the problem. So, really, the emphasis was not on shifting away from inspection totally. But, rather, recognizing the need for more proactive approaches to balance that. The auditing or inspecting of existing structures generally only tends to attack the symptoms (such as particular defects) and not the root causes (why those defects are being introduced). The emphasis, from Deming's point of view, was a process dedicated to fixing defects rather than a process dedicated to improving the means by which to eliminate defects that require being fixed in the first place.
Deming says: "Eliminate the need for inspection on a mass basis by building quality into the product in the first place."
Think of "inspection on a mass basis" as being a reactive testing only process. As we all know, we want quality testing to be more proactive, by testing at all elements of the product development lifecycle. Many software places feel you can get by just with testing after the whole application is put together, just as many manufacturing firms thought that it was possible to build and then inspect after the building was finished. Also remember that at earlier times in our own industry there were "mass production" tool-types for producing software, such as certain types of CASE tools. This led to a certain type of "mass inspection" process that had a lot of problems because it did what manfacturing did: it made the "assembly line" the testing ground. That is not good. If you wait to the assembly line to do sampling, inspection, auditing, etc., then you are waiting too late in the process.
So the whole focus on this Point 3 is really talking about the nonquality costs of rework and scrap in terms of how the product is produced and it is a polemic for a more proactive approach to quality such that as much quality as possible is attempted to be designed into the product, rather than being "tested" into the product after the fact. It is trying to focus people on creating more upstream processes rather than relying on the viability and veracity of processes downstream.
Re: Deming\'s 3rd Point
You are greate. But I will thank you only after 14th point.
Sorry for that!!!
[This message has been edited by AmitK (edited 02-12-2002).]