“Halting the Line” Improves Quality

Well conceived and implemented processes are essential to the delivery of consistently high quality products and services.  Much can be learned from Toyota and its Toyota Production System.  One of the pillars of the Toyota Production System is the commitment to quality control throughout the manufacturing process, rather than reliance on inspections after a vehicle has already been produced.  Called “jidoka,” this principle grants all front level employees the authority to stop the production line in case of abnormalities.  According to Jeffrey Liker and David Meier in their book “The Toyota Way Fieldbook,” a continuous and system-wide focus on quality control using the principle of jidoka is one of the key reasons that Toyota’s manufacturing process has achieved consistently superior results compared to other automobile manufacturers.

Jidoka can be applied just as effectively in a service process, such as large scale document review.  The currently prevailing process of conducting large scale document review rarely permits front level reviewers and their immediate supervisors to halt a review pending resolution of error causing ambiguities.  Often, reviewers fail even to notify supervisors of ambiguities, choosing instead to make a partially arbitrary decision and continuing to the next document.  Even when reviewers ask questions to supervisors, the subject matter of the question and the supervisor’s clarifying instruction are rarely communicated to the rest of the review team.  And this is true for reviews conducted solely in the U.S.  If the LPO scenario is considered, the time zone difference between the review team and supervisors prevents an immediate response to error causing ambiguities. 

 

The traditional large scale document review process ignores the true value of jidoka – when a Toyota employee halts the production line to address an abnormality, the end result is not just correction of the error for that one vehicle, but also prevention of the same type of error for future vehicles coming down the line.  At Aphelion, our document review process is designed with jidoka in mind.  Our review team is trained to identify abnormalities in the document set – essentially documents where additional information or clarification is necessary to make the correct decision.  Once a question arises, a review team member must immediately bring it to the attention of a supervisor.  For questions with broader implications for the review, the supervisor will immediately halt the review to provide instruction to the review team, so that other documents in the same category are designated correctly. 

 

Loss of time is the obvious concern with a jidoka driven process.  By demanding that our front level reviewers take action that may “halt the review,” we run the risk of unnecessarily delaying the review.  But this concern can be addressed with proper planning and the presence of an on-site U.S. licensed supervisor with the experience and legal acumen to take swift remedial action to minimize any delay.  Toyota’s manufacturing employees have the skill and experience to exercise their judgment on something quite complicated – building a functioning car.  Aphelion hires and trains its Indian review team so that our reviewers have the necessary tools to exercise proper judgment in large scale reviews for complex civil litigation matters.  Our employees possess the skill and judgment to determine how to designate a document properly and to assess when a document presents issues that are not adequately addressed in the review memorandum or initial project training.  It is our principle, similar to Toyota’s, that if a document review process does not demand the front level review team to “halt the review” when necessary, a significant portion of the human capital available will be wasted and quality will suffer.

One Comment

  1. jwaterman:

    Excellent piece, Hiren. The amount of ambiguity and inconsistency on the average doc review is considerable. Most of the information flow is downstream. And as you nicely summarize, the little feedback making its way back upstream does not get circulated back down as it should. In my experience, most reviewers try very hard to be consistent and accurate in their work. Consistency between reviewers, then, becomes the bigger problem for all the reasons you outline above.

Leave a comment

You must be logged in to post a comment.