A glaring weakness in the "lean" community (lean manufacturing, etc.) can be addressed with a new perspective, a new approach, and our technological breakthrough.  The weakness is the lack of a data-driven, logical lean implementation system for huge multi-national, multi-business corporations.  Over the last six years at Virginia Tech, two things have bothered me about modern lean production implementations: (1) why are major corporate implementation failure rates so high, or at least why are they not more often huge successes? and (2) how can a massive multi-facility, multi-business, multi-national organization integrate lean practices in such a way that the organization reaps the maximum benefit?

Over that time I’ve come to the conclusion that the state-of-the-art in lean technology isn’t effective enough for a conglomerate implementation.  Even the best lean practitioner is limited by the cognitive capacity of being human.  The good ones can walk into a facility and see plenty of low hanging fruit to improve the system.  The great ones can observe a few facilities and begin to lean the supply chain between them.  But in the modern world with globalization, the next step is to enable the conglomerate to roll together the lowest level value streams into a corporate-wide high-level value stream.  This is far too complex to leave to human guesswork, and requires a significant technological capability and data-driven, rather than experience-driven, decision-making.  This is why we started building our lean software package in 2005.

Our Industrial Engineering group is building on these ideas that were formed over the last couple of years, and addressing this weakness with a new approach.  We call it ImpactLean.  The system incorporates new technology to use data to predict and quantify the impact of a lean improvement.  The technology is able to roll low-level value streams into a high-level corporate value stream, accurately representing a system of systems.  I think it is a significant breakthrough, and we're pretty close to proving it.  If you're interested in hearing about it, let me know.