Analyzing Field Failures to Improve Manufacturing Quality Control

It’s not always easy for businesses to maintain high product quality and customer loyalty without sacrificing profits. When it comes to manufacturing and maintaining profitability in the face of everything threatening a business, companies are often too quick to cut corners on quality, when they could instead take steps to prevent defects and returns.  There are three factors that should be established with the integrator around expected yield/ failure rate: Major/Minor, Failure Trends and Root Cause Analysis. Each of these factors is important to your product’s expected yield for defective systems/machine return rate. Now, many of you are thinking, “I don’t have a product failure problem,” but for some products even having upwards of five returns in consecutive months with same symptoms, is a signal there is are issues with returns that will eat into a company’s profits.

Major/ Minor

The first step is to make a list of reasons for product being returned to the manufacturer. Then define which issues you consider Major or Minor and separate the two. Typically, cosmetic errors like crooked labels and scratches are considered minor issues. Where as being unable to power up a product, excessive damage caused in the packaging process, blue screens, system hangs, or missing parts might be considered major.

Failure Trends

The next question to ask is are you seeing any trends with the failures you are experiencing?  If you have more than two storage servers and the video conferencing systems that you recently installed are showing signs of failure, then you have the potential trend for a high risk of component failures that will need to be addressed immediately for the possibility of further problems. In some cases, you may have to recall potential systems back.

Root cause Analysis

The next step, which should come from the manufacturer, is creating a detailed step-by-step failure analysis report that can help you and your supplier nail down what is causing the system to fail and in what stages of production, so they can be corrected. Early stage, latent failure,  like incompatible components, ESD handling process, improper instructions/training, qualification of the production associated should all be recorded.  Symptoms like an overstressed motherboard catching fire fire on the field, or a loose screw getting stuck inside a system fan and breaking it, for example, can be attributed to improperly calibrated tool use.  After root cause is determined, manufacturing should be able to ascertain an appropriate fix as well as what actions are required to stop the trend  in the future (containment/ corrective/ preventive action). Example: If the root cause is a power overload that is causing capacitors from the power supply to become exposed, the manufacturer should be capable of issuing your company a quality alert for a possible recall. This alert should include your sub supplier for defective products (even they have not failed yet), so they can stop shipping immediately and segregate all defective products. Corrective action would then be taken by the sub-suppliers who manufacture your power supply.  A 3rd level root cause analysis should then be conducted (with why method). Most good manufacturers have processes like this built in, but if not, it is imperative that your supplier be able to provide you with both short term and long term corrective actions to failure trends in order to ensure the cost you are incurring stops at the point of discovery. Quality Assurance + PC Manufacturing