Manufacturing Excellence:Reducing Cost of Failures
January 2, 2008
Sustaining profitable growth must be the mandate for every business in order to create opportunities for employees, value for stakeholders and contributions to the community. Without profit, growth is borrowed, and without growth, the profit margin shrinks. However, many manufacturing businesses suffer from limited profit margins because of lower yields and excessive waste.
Low yields, poor process flows, excessive inventories and inefficient management systems all contribute to the loss of profits and job loss. But the high failure rate leads to consequential wasteful activities. Product or service failure rate is the key differentiator between good and bad companies.
My experience has shown that the difference between a superior and the average company may be a 10% to 40% difference in yields. Even a 3% to 5% extra loss in yield can create a competitive disadvantage because loss of yields and cost of handling the loss directly affects the bottom line.
Almost every company has a quality improvement program such as statistical process control, design of experiments, lean or Six Sigma. Sometimes, these quality improvement programs cost more than the improvement they drive because of incorrect application without paying attention to the intent of the method. Most process improvement teams look for the root cause on the surface and the symptom level. As a result, the corrective action is ineffective and poor performance perpetuates. The process outputs are continually inspected, verified and reviewed.
We have learned that instead of performing root cause analysis and taking corrective action, if one spends a little extra effort in setting up the process correctly, it pays huge dividends.
In order to set up a good process, one must create a sense of preparing for the process that involves ensuring good material, method, machine and employee training for achieving the defined target performance.
We all know the saying, garbage in garbage out, but we still do not ensure good inputs. Planning for the quantities to be produced, as taught by the PDCA (plan-do-check-act) model, is not sufficient. Instead good preparation is required to produce virtually perfect output.
Spending another dollar in setup can lead to savings of multiple dollars in production, and contribute ‘big’ to the corporate profit.
There is a caveat, though, to having good preparation. The design engineering team must specify target values for setting up the process. The setup cannot be acceptable-it must be virtually perfect, that is, on target. The challenge setup technicians face is that prints, drawings or procedures define range of values for setup, thereby creating fuzzy targets leading to shoddy products, loss of customers and loss of jobs.
A small electronics manufacturer practiced the methodology in documenting procedures for their quality management system. They have been able to improve productivity about 3% to 5%. A testing process at University of Military Intelligence deployed better preparation ensuring the necessary material, information, tools, machines, methods and skills were available. More than 2,000 people hours were saved at the testing process. Again, we all know that good preparation pays dividends but we fail or forget to practice it in favor of rushing to finish work and having the mindset of ‘just do it.’
Thus, it is the responsibility of each and every employee to prepare well for one’s process, be it production or support. To prepare for a task at hand, it is important to ask or know what is needed-which should be captured in processes documentation or work instructions-so one can modify the procedures to ensure preparation is spelled out. Conventionally, the preparation has been left out in order to rush to get to the job to be done.
We all have seen chaos in absence of improper preparation or the panic to meet the process output demand in absence of process problems.
Try out your process yourself. Take the current process and add the necessary steps to ensure the good preparation to run the process virtually perfectly.