Like any manufacturing plant, CNH Wichita Product Center aims to keep its quality yield high and its scrap and rework low.
Data show the progress it has made: The plant, a 2012 IndustryWeek Best Plants winner, reduced its scrap and rework costs by nearly 81% over a three-year period, a significant achievement by any measure.
Helping to drive CNH’s quality improvements is a robust quality process that includes quality “gates” and a quality “network.”
Here is how the process works. It begins with the QA matrix, which is exactly what it sounds like: a matrix that outlines any and all quality-related problems discovered at the facility. Even more, the quality issues are ranked and stratified based on a variety of parameters, such as frequency of occurrence, cost of labor involved in rework, and more.
A review of the matrix determines where a quality “gate” may prove helpful. For example, CNH Wichita Product Center, which makes skid steer loaders and compact tract loaders, determined that an ongoing quality issue on a line warranted the introduction of a quality gate on the line. At the gate location, a team member examines product as it reaches the gate against a checklist of factors. If a defect is found, the team lead replaces the operator who performed the operation. That operator then is freed to examine the error, correct it and then verify that no additional defects were introduced between the point of operation and the quality gate.
“If the same defect is found twice, then the supervisor and/or team leader are required to perform a HERCA [human error root cause analysis]. This is a series of question that help define if the root cause of the defect is a man-related issue or a process-related issue. We then have specific tools that we use to address both root causes,” explains Garry Davenport, quality manager.
In short, the quality gate is a step in the process in which data are captured, work is reviewed and action is taken if the review shows it is needed. Quality gates can be removed once no defects have been detected in a 30-day period. Moreover, Davenport says, a robust fix must be implemented, mistake-proofing techniques introduced and sustainment validated.
The process works. In the line identified in the previous example, the introduction of the quality gate drove down the defect by 97%.
The quality “network” provides another means of identifying and correcting defects “as close to the source as possible.” Briefly, the network calls for workers on the line to inspect one or two key components on a unit as it reaches them to assure the previous task was performed correctly. The aim: To find and fix problems before they are hidden by a process further down the line.
