HomeContinuous versus Batch Flow
The holy grail of a process is to achieve continuous flow where the transfer quantity between successive processes is one piece (one piece flow). Prior to lean manufacturing, many plants were "batch" organized by department or function. e.g. In a foundry, there would be the following departments: tool and die storage, core making, moulding, shake out, melting, core knockout, de-gating, profile grinding, shot blasting heat treat and inspection. In such an operation, lead time can be as high as six months as the product is moved, stored, processed stored and moved many times. Often the product is lost and must be cast and processed a second time. Not surprisingly, customers are very unhappy whith such lead times and will change vendors to get a shorter lead time.
Linking these processes into a cell and coordinating the core making, melting and moulding schedules can reduce lead time to less than a week. For example, in the above foundry the automatic moulding machine can be used to clock the whole operation forcing all other processes to run at the same speed. In such operations, quality problems show up immediately and the process is stopped proactively with a trivial amount of waste. In a traditional shop organized along functional lines, quality problems can exist undetected for years at great cost.
There are a few caveats or cautions to the goal of achieving continuous flow:
1. In a service repair shop where there is no inherent flow between manufacturing repair processes (cleaning, welding, machining, coating and de-burring) and assembly and test, any attempt to achieve flow in such an operation will be a complete waste. A large pump OEM company tried to implement lean and process flow in their 50 repair shops world wide. After two years of achieving only minor improvements, the initiative was abandoned and the staff lean consultants were reassigned.
2. Automated machining transfer lines offer up another caution. A very large automotive OEM company installed a V 8 block transfer line in a Canadian plant in the late 1990's. With enthusiasm to be seen to be implementing lean manufacturing, the line was designed with one piece transfer and did not contain any storage buffers between subsequent machine tools. The actual throughput achieved was a fraction of the intended. The engineers did not consider that a typical machine tool may only run at 60% up time due to tool changes and part inspections. With several machine tools on the line, the expected throughput (up time) is 0.6 raised to the seventh power, in this case 3%. Subsequently this line was re-engineered with a scientific assessment of the needed buffer sizes between each machine tool. After refitting the line with appropriately sized between process storage buffers, it proceeded to run at its intended throughput.
Burns Bridge can help you with your process flow improvements.
