Execution: 

• A preliminary assessment of the current state showed that two actions were required.  The first was to reduce the variation in the eight process sub steps and to reduce their mean cycle time. The second was to eliminate schedule delays between sub steps in order to create a 12-hour maintenance window. This would be a total reduction of 24 hours or 40%.
• Steve directed a multi-disciplinary team using DMAIC variation reduction and detailed analysis of each process step.  The process was divided into segments and staff were assigned to improve each of the eight sub-steps.  
• Each sub-step was analyzed with value stream mapping and time study.  Since there were 4 different shift teams that performed this operation leading to a large variation in cycle time; so standard work was applied to define the best practices.  We wanted to make the work highly repeatable.
• In many cases we eliminated “white space” between sub steps caused when staff left to work on a different process. We established that once this process started, each sub step would be performed aback to back. 
• Some of the general improvements involved reducing waiting time for material delivery and for test results.  In some cases we found that cooling and heating times could be improved.
• The results were reported weekly, graphed, discussed in a team meeting and improvement steps were implemented. We found that some steps could be done in parallel instead of in series, helping cycle time. 
• Several lean tools were used to make the necessary improvements including 5S, standard work, reduction of motion and waiting waste, 5 Whys, Quick Changeover, spaghetti diagrams, cell design and time studies. To prioritize improvement recommendations, we developed a 9 cell chart comparing the benefit os proposed changes to the expected difficulty level to implement them.