TOCICO

Goldratt asks if the audience wants him to evaluate MLIT’s use of critical chain with contractors. Many in the audience wanted the truth. Eli’s staff wanted politeness. Eli decides to be critical of MLIT’s use of critical chain. Project managers should have seen a 25% reduction in project lead time. If you think that by using critical chain software you are doing critical chain then you are wrong. Do you believe by the government committing to a one-day response to contractors' questions and taking one week is not critical chain. Some responses should be given immediately and others are not so important. What are the criteria for evaluation of the response? If the project will be delayed because of the response delay then answer immediately; it is important. If the delay has no importance in its relationship to the project completion times, then why the rush? There are three important steps in projects. First, choose the right project. Second, design the project correctly. Third, execute the project correctly. For the first question, the thinking processes (analyze the situation to determine the real need—that is that the project meets that need) are used. For the second question, the design of the project, the tool is the S&T tree, how do you start with an objective and end up with a project network? The whole logic is spelled out. How can you use the gemba experience if others don't know why you have designed the project in the way you have designed it. For the execution you start with the PERT then end up with reality. Your biggest waste (according to Ohno) is overproduction. In project management the biggest waste is buffering each task. Critical chain says to take this local buffering and protect the whole project. By aggregation of local buffers the variability goes down hence project lead time must go down by 25%. If you don't report on it then you do not use critical chain.