Cook-Hauptman Associates, Inc.

Step Change Notes from First Factory Meeting

By Jim Cook ( 1990 )





Step Change

Our Focus

¯ )


In February, I was engaged to consult on improving Fibers' R&D allocation process. Interviews within the Fibers' Department revealed a consensus for Step Change improvements in the Quality, Responsiveness, and Availability of our products and services, and for improving our relationships and returns on capital. Other revelations, however, seemed at odds with the consensus, specifically:

  1. Management's bias toward short term predictable results and measurement using traditional cost accounting,
  2. Management's bias toward consistent performers over erratic performers,
  3. Research's bias towards specialization, introspection, skepticism, and scale, and
  4. Inconsistencies between Operation's needs and Research's pursuits.

Although there was a universal belief in Fiber's technical capability, there was widespread apprehension about future prospects.

Upon reflection, we surmised that the Department (and its Divisions) did not have:

  1. A shared vision of the business context and the Department's (and the Divisions') role(s) for the decade of the '90s and
  2. A Technical Agenda in support of business strategies on which to build a constancy of purpose for the decade of the '90s.

We concluded that the increasing difficulty in achieving business results in Fibers' R&D is due to not breaking out from the past. Whereas the customers' reality (in terms of options open to them and demands placed on them) is undergoing a Step Change in Quality, Responsiveness and Availability (due to Pacific Rim capacity and the Global utilization of information technology), Du Pont's offerings are not keeping pace. Whereas bargaining power is shifting from the supplier to the customer, Du Pont is not as responsive as customers would like.

And ... that brings us to our intent. Du Pont, Fibers, and Fibers' R&D need Step Changes. But to just dictate Step Change Strategies and Agendas would be suicidal, and everyone knows or senses that. What Du Pont Fibers needs are Step Change Exemplars in order to have a methodology and process for doing Step Changes. Step Changes are not new, not even for Du Pont. What is new, and is urgently needed, is a process for doing Step Change.

The push for improved First Pass Yields offers a splendid opportunity for us to participate in a group or groups with a view of offering external experiences and scoping a process or methodology. In turn, with a Step Change process or methodology, Fibers' can confidently adopt Step Change Business strategies and R&D can have a Step Change Technical Agenda in support of the Businesses.

What follows is the outline of the discussion at Camden on Tuesday, August 14 from 8:30 to 11:45 a.m.



A. The purposes at the Learning Level are to:

  1. Demonstrate to the Department that Step Changes have commonalities in their structure, methods, metrics and results that enable new challenges to be tackled more systematically.
  2. Provide tools which enable Step Change to be started and pursued confidently.
  3. Contribute to the building of Departmental confidence and esprit regarding Step Change.

B. The purposes at the Strategic Level are to:

  1. Provide a solid foundation for lights-out, high flexibility, high thruput initiatives.
  2. Contribute cash flow improvement to the Department.
  3. Enable the development of a Technical Agenda which enables Production and Sales (and Service) to perform in a marketplace experiencing and demanding Step Changes.

C. The Tactical Objectives are to:

  1. Achieve a significant reduction in rework and waste.
  2. Improve understanding of processes through interdisciplinary, surgical, use of collaborating resources.
  3. Demonstrate repeatability of Step Change methodology.

D. Relation to other World Class Manufacturing initiatives:

  1. To complement by building upon (not contending for substitution).
  2. To become another World Class methodology for use by Du Pont Manufacturing.

E. Today our purposes are to:

  1. Share views, ideas, experiences, and observations regarding Step Change.
  2. Develop a consensus which views going ahead as worthwhile.
  3. Collaboratively prescribe steps and identify candidate problems, people, structure, and resources.


A. Beginning last February, the Technical Directors, Tony Cardinal, John Fallon, Ray Johnson, and others were interviewed regarding Technical's Resource Allocation Process and results.

B. Several messages emerged:

  1. Global Markets were undergoing Step Changes (from Scarcity ---> Surplus, Simplicity ---> Complexity, Stability ---> Volatility).
  2. Step Changes in Fibers business conduct is vital and are threatened by some traditional practices and forces.
  3. Fibers is unsure about how to go about Step Changes and therefore is unsure about the Future.

C. We surmised that:

  1. The Resource Allocation process should be done after development of a Fibers R&D Agenda which supports the business needs of Sales and Manufacturing in this rapidly changing Global environment.
  2. Such an Agenda will undoubtedly require a Step Change in how Fibers R&D is conducted and will not come to pass until the Department has more confidence in its ability to do Step Change.
  3. That confidence will come about when there is an Exemplar instance of Step Change and a generalized methodology for doing Step Change.

D. The necessary ingredients for solving the Step Change dilemma exist:

  1. Compelling Need: Manufacturing's focus is on Step Change in First Pass Yield.
  2. New Technology: History shows in case after case that new technology is a necessary enabler.
  3. Near Term Feedback: In addition to Strategic benefits, specific measures of success are achievable within a year.
  4. The above presumes:
    1. Inspired participants emerge and are resourced,
    2. The essential management empowerments evolve, and
    3. We tackle a suitable problem effectively.


A. What we mean by Step Change is:

  1. Greater than 2:1 gain in performance within 12 months.
  2. By a systemic attack built upon an insightful strategy.
  3. With visible, meaningful business results.

B. We can cite as examples:

  1. Beretta (the pistol manufacturer) who went through 5 manufacturing process revolutions, each with a 3:1 improvement in productivity and invariably accompanied by a concomitant quality improvement.
  2. Personal examples drawn from the computer and electronics field.
  3. Du Pont's ETF experience in the late '50s when a joint undertaking by Engineering and Textile Fibers collaborated to create the "optimum nylon plant."

C. These examples offer the following instructions:

  1. The gains are most often around 3:1, although that can vary widely, it is always greater than 2:1.
  2. A unifying principle ensures that parallel initiatives will have additive effect and de facto coordination.
  3. A particular new (to the context, but often otherwise old) technology seems to be pivotally leveraged.
  4. A reconfiguration of roles, power, and knowledge seems to accompany and follow from the results.
  5. In many instances, mult-functional collaboration (Mfg. and R&D, Sales and R&D, Engr. and Technical) was central to the success.


A. The benefits of achieving high FPY to:

  1. Manufacturing are:
    1. Material savings: less scrap and rework.
    2. Space savings: less storage and rework space.
    3. Expenses savings: rework labor and disposal expenses.
    4. Process simplifier and automation enabler.

  2. Sales are:
    1. Higher customer satisfaction.
    2. Easier segue into trusted and preferred supplier status.
    3. Fewer service write-offs.
    4. Better customer relationships.

  3. Technical/R&D are:
    1. Knowledge enhancer.
    2. Process idea generator.
    3. Product idea generator.
    4. Respect generator and relevance dsplayer.

  4. Du Pont, generally, are:
    1. Higher personal satisfaction by everyone involved.
    2. Enhance image (and reality) of quality.
    3. Better margins, cash flow, and Return on Net Assets.
    4. Higher opportunity.
    5. More value to share with customers.

B. Why FPY? (or, why not Thruput, Lights-out, ... ? ) :

  1. Because of two priortization rules:
    1. Within a system, every solution to one problem is a constraint on the entire system (to the extent that the solution interacts with the other parts of the system).
    2. A corollary is, often a solution to a global optimization (such as cost minimization) is the implementation of a local solution which is sub-optimal, but whose system wide interactions actually effect a global optimization.

  2. Translation of the above two rules reads:
    1. Solve the tough problems first otherwise the solutions to prematurely solved easy problems will get in the way of (and sometimes preclude) solving the tough problems.
    2. Keep the global system goals in mind (satisfying customer desires profitably) and allow (i.e., don't rule out) reconfiguration of some local costs.

  3. Applying the above:
    1. Specifically, before going for ultra-thruput or lights-out (i.e., ultra-automation), we need to solve the tough problem of near-perfect yield).
    2. Specifically, rather than further optimization of the parts (i.e., locally), we need to adopt a strategy and apply it even though it will likely increase the cost of some part of our production system.

  4. As with the Step Change examples, we have new technology (i.e., real-time, on-line computerization and off-line computer modeling) which have barely been applied to Du Pont's plant operations, especially the older ones.

  5. There is a high level (Ray Johnson, VP Manufacturing) push for improvement in First Pass Yield, and the Departments' projected financial performance can use improvement.

  6. First Pass Yield improvements are achievable, have been achieved by some (Japanese) competitors, and have been achieved consistently over the past 25 years by the semi-conductor industry.

C. The discussion was lively throughout. A particular point of discussion centered around local optimization versus global optimization. "Should we tackle the entire BCF plant or optimize something local, like elimination of breaks?" "Doesn't optimizing locally run counter to the second prioritization rule?" (It does, however, if the sub-system has been architected to minimize the interactions with other sub-systems, as "good" architecture tries, then that architecture enables sub-system optimization, without high overall system costs.) The breaks were felt to be suitable candidates for Step Change reduction. That was the consensus, not the conclusion. We did conclude that FPY is an appropriate problem based on the argument that the benefits are great and it is necessary to have high FPY before doing high thruput, etc.



HOME E-MAIL as of January 20, 1998

Copyright © 1990 by James E. Cook