In my auditing travels, one practice that I see on a regular basis that makes no sense is the practice of measuring supplier performance. Specifically combining objective measures and subjective measures and trying to calculate a formula for acceptance.  Let’s get one thing straight.  It cannot work!  Let me explain the typical story.

Companies in their attempt to meet ISO or TS 16949 requirements develop a process that gives scores for Quality and Delivery.  They then try to combine these scores in order to have one quantifiable number for supplier performance.  In these processes it is typical for a supplier to start with a number of points and to lose them for poor performance.  For instance a supplier has 40% of there score for quality.  If they have 1 rejected lot they lose 10% and there score is now 30% etc.

The same method applies for delivery.  Again, typically the supplier must maintain a score of 70 or 80%.  This in itself is usually unworkable since a supplier can have 2 rejects per month and maintain an approved supplier condition.  But companies don’t stop here.  They then add to the formula 10% for having ISO certification and 10% for cost reductions etc.  Now the acceptance which in theory is 100% becomes 90% because we are not really going to penalize anyone for not constantly providing cost savings.   The 10% for ISO certification is automatic if a company is certified and disregarded if they are not since we keep using them anyway.  So now the real score is actually out of 100, but everyone has 20% automatically.

Therefore if any supplier has one delivery issue and one quality issue every month they are considered acceptable with an 80% score.  Often in reviewing these programs I can point out to the owner of the supplier performance process that the supplier which he identifies as the organization’s worst supplier is actually identified as acceptable in their scoring system.

So let’s get to the point.  These formula based scoring systems simply don’t work.  There are too many combinations of conditions that need to be taken into consideration and the formula that would accurately address each situation is far beyond the math skills of most of us.  Plus the practice of combining measures of quality and delivery with cost issues and systems issues is comparable to adding apples and oranges.

So what works?  The simplest answer is to score each issue separately.  Score quality based on the quality needs of the organization.  100 % good product is acceptable 1 reject is over a defined period is of concern and 2 rejects is unacceptable period.  Likewise 100% delivery is the requirement however 1 late delivery is acceptable but requires attention.  Two late deliveries over a defined time frame are not acceptable.  Each requirement is a separate unique situation.  If the organization chooses to add other conditions then they are measured separately and do not effect any other requirement.

The expectation of the organization then becomes the achievement of each requirement on its own merit.  Any one condition which is not met requires appropriate corrective action regardless of the performance related to other requirements.  Often this process evolves into a simple ‘RED’ ‘Yellow’ ‘Green’ stop light system.

Consider your organization’s system for supplier performance tracking.  Is it effective?