Peter UlintzThe Reality of Problem Solving
July 25, 2022Comments
Metal forming problems can be very difficult to solve, as determining the root cause often proves impossible. Problem solving becomes more difficult when we make decisions based on what we see and believe is true about the problem at hand.
Let’s address these three statements individually.
Metal forming problems can be difficult to solve. Metal forming involves a collection of process inputs that may include machine selection and its condition; tool design, construction and maintenance; lubrication type, mixing practice and application; sheet metal properties and ordering practices; environmental issues (process-related vibrations, operator inputs, deformation temperatures, thermal expansion, etc.); and other factors. Because process inputs can and do change over time—sometimes in a matter of seconds—they also are called variables. The prevailing combination of input variables interacts to produce a process output. In metal forming, the desired output is a product that meets print specifications, and that the customer will accept.
For comparison’s sake, consider a four-tumbler padlock (Fig. 1). Each tumbler contains 10 numbers ranging from zero to nine. When the numbers on all four tumblers match a very specific sequence, the lock opens. Only one combination opens the lock. But how many combinations are possible? Because each of four tumblers offers 10 choices, the possible combinations number 104, or 10,000.
So Many Variables
Now imagine that each tumbler on the padlock represents a process input: machine condition and maintenance, sheet metal thickness and properties, tooling design and maintenance, and lubricant concentration and film thickness (Fig. 2). Each number on the tumbler represents a single input variable. Machine variables for a stamping press may include counterbalance pressure, running clearances, thermal expansion, ram parallelism, gib clearance, among others. Variables in sheet metal include thickness, yield strength, tensile strength, total elongation, etc.
We easily can find 10 or more variables for each process input in this example, which means that 10,000 variables or more influence the process (stamping) output. But process inputs number more than four (straightening, feeding, material movement, speeds, pressures, etc.), meaning that perhaps more than one million variables can—and do—affect the process output.
Many manufacturers view root-cause analysis as a quality-assurance function. Therefore, they do not apply it to complex metal forming processes. Nor do they apply even simple problem-solving techniques such as five whys, fishbone diagrams and Pareto analysis to solve and document metal forming and die-related problems.
