It’s my opinion that many people think solving problems has to be complicated and time consuming. There is always the thought in the back of their mind that they might fail and be looked down upon by their boss and peers. It is for these reasons (and others) that many problems become part of process waste and inspection is used to identify and remove them. I would like to discuss two approaches to solving problems, divergence and convergence.
A divergent process is one in which a team brainstorms the different variables that affect the process. They could use a fishbone diagram and start with the 6Ms, i.e., Man, Machine, Material, Method, Measurement, and Mother Nature. They identify the different variables associated with each of the Ms and put corrective action in place to reduce the associated variation. Using this approach, significant process improvement can be made.
A convergent problem solving process, on the other hand, focuses on identifying the dominant cause of the problem. It uses process of elimination to step your way through the problem. Various convergent tools are used that eventually give you clues and ultimately lead you to the dominant cause. One of the most simple and powerful techniques is comparing extremes. What is different between the best sample and the worst sample? Best day and worst day? Best machine and worst machine? Etc.
Another technique is segmenting or stratifying the data. You might start by graphing all the data. This will give you some sense of where the data is centered, the shape, and the dispersion or spread of it. However, to make any headway you will need to break the data up by machine, shift, person, etc. to help identify where the dominant cause lies.
In future postings, I will discuss some of the other simple techniques that can be used in solving problems using the divergent or convergent process. As always please feel free to add any comments you may have concerning the above or any real life experiences solving problems.