8.1. Setting the Scene

For the past 25 years, MoldMat Ltd. has supplied the plastic used by one of its major customers in Italy in making white garden furniture. Over the years, mysterious crises occur during which the flowability of the plastic leads to low yields for both MoldMat and its Italian customer. To date, all efforts to find the root causes of these crises have failed. After the most recent crisis, a new team is formed, led by an engineer named Carl Linton, a black belt with training in Visual Six Sigma.

8.1.1. Manufacturing Process

White garden chairs and furniture command a very good price, but they are difficult to make owing to the impact of whitening agents on plastic flowability. Getting the right balance of whiteness and flow is not easy. As the proportion of additives in the mix increases to make the plastic whiter, the flow of the plastic is impaired.

The process for making white plastic begins with the preparation of a filler mixture, or slurry. The white filler, which is an inert powder, is sourced from a number of quarries in Africa. It is mixed with unpurified river water in a stirred tank in the filler preparation section of the MoldMat polymer plant (Exhibit 8.2). The filler preparation tank is agitated and held at a target concentration. The tank is topped off each day with filler and water. Small amounts of a viscosity modifier are added to the slurry if the viscosity gets too high.

Clear plastic is made by heating and stirring a monomer in a batch reactor until it polymerizes. To make the white plastic, the filler slurry is added to the monomer in the polymerization reactor at the start of the polymerization process. When the polymerization reaction is complete, the molten polymer is granulated and packed. The MoldMat plant in England makes three batches of white plastic per day, running a 24-hour schedule every day of the week.

Figure 8.2. White Polymer Manufacturing Process

The polymer plant tests every batch of polymer. A sample from each completed batch is taken and tested for:

• Color (whiteness), measured on a colorimeter using a color index.

• Melt flow, measured as a melt flow index in an off-line laboratory test. This is an indicator of how well the polymer will process in the downstream molding plant.

• Filler content of the polymer.

8.1.2. Typical Crisis

Crises have occurred two or three times a year ever since a new product was introduced ten years ago. Here is a typical sequence of events.

The Italian molding plant will have several months of normal processing before starting to experience problems with flowability. When this happens, technicians in the molding plant check the processing parameters, and if these look reasonable they question the quality of the polymer. The MoldMat plant engineers check that the polymer is in specification and verify that there is nothing wrong with the test equipment. This leads the processing plant engineers to suspect that the molding processing parameters have changed.

After a few more days of bad processing, the molding plant engineers will ask for some different polymer to run as a trial. This requires a fresh start for molding production. The molding plant must empty the polymer silos to run the trial polymer. The purged material is sold as scrap, which is accounted for as a loss in the MoldMat plant yield.

By this time, the output of the molding plant is well behind schedule, and customers are running out of chairs. The business suffers substantial lost margin and opportunity.

Meanwhile, rapid action teams have been assembled from across Europe. A plethora of helpful theories and their associated solutions are developed, such as:

• The filler supplier is inconsistent and should be replaced.

• Last week's heavy rain has altered the pH of the water supply, which has affected the reaction chemistry.

• The MFI specification is too high, so batches of polymer at the bottom end of the specification range should be the only ones used.

• Abnormal ambient temperatures and humidity are to blame.

• The filler is not evenly distributed through the polymer, and agglomerates are blocking the flow channels in the molds.

Process changes are made, trials are run, and data are gathered. But none of the changes ever conclusively solve the problem.

Then, mysteriously, the problem goes away. The molding process gradually improves with everyone convinced that their pet theory or solution was the one that made the difference. All is well until the next time.

8.1.3. Forming a Team

After one particularly bad crisis, the manufacturing director, Edward Constant, has finally had enough. MoldMat has started to implement Visual Six Sigma, and the black belts from the first wave of training are anxious to start driving improvement. Edward is skeptical about Visual Six Sigma, but he is prepared to give it a go—after all, nothing else has worked.

Carl Linton, a bright young engineer who has only recently moved to the polymer plant, is one of the first trainees. Edward has met him a few times and is impressed by his openness to new ideas and his approach to problem solving. Given the numerous false starts, Edward figures that Carl's lack of detailed knowledge of MoldMat's operations could actually be an advantage, provided that he works with people who have the right mix of experience. Edward agrees to act as the project sponsor.

At their first meeting, Edward tells Carl that he will give him all the help he needs. "Everyone has an opinion on the best solution, but I have never been satisfied that anyone has properly done any rigorous analysis, let alone identified the root cause of the problem so that it can be conclusively fixed," he says. "This problem has been around for ten years, so a few more months are not going to make that much difference. The best advice I can give you is to take your time and to trust nothing and no one, unless you have personally verified the data and have worked through it in a methodical way. I don't want any more crises. If the process can work most of the time, then it should be able to work all of the time."

Edward knows that a change of polymer can immediately affect the processing performance of the molding plant, even if the polymer batches meet the polymer specifications. So he directs Carl to focus on the polymer plant first. He urges him to talk to a wide range of people in both the polymer and molding plants. But above all, he directs Carl to collect some data.

Carl and Edward decide to form a small project team comprised of Carl and the following associates:

• Henry, the polymer plant quality manager.

• Bill, a polymer chemist from a technical support group.

• Roberto, a process engineer from the Italian molding plant.

To ensure that the Visual Six Sigma methodology and tools are correctly applied, Tom, a well-seasoned and culturally savvy master black belt from MoldMat's training partner, supports Carl. Carl and Tom assemble the team and review its objectives. To ensure that all team members share a common language and approach, Tom and Carl schedule and conduct an impromptu training session in Visual Six Sigma and JMP.