At Setpoint, we see a lot of remarkable product ideas come across our table. These products range anywhere from lip balm to aerial vehicle decoy flares, snowboards to bullet shell casings. Developing, testing, and building a lean process to mass produce these products is our bread and butter here at Setpoint. During our years of working with brilliant and sophisticated customers we have found a common problem that undermines many of their great product ideas. Though these products have great potential and even in some cases, strong appeal to my personal wants (snowboards), they suffer from a common disease – NDFM. They are Not Designed for Manufacturability.
Designing for manufacturability (DFM) is the process of proactively designing products for optimization during fabrication, assembly, purchasing and testing. Other factors include making sure the product meets third party regulations and shipping constraints. For my purposes here I want to focus on the levels of fabrication and assembly.
The process of designing for manufacturability should happen before the product idea is used as a foundation for building a product empire. If there’s a critical crack in the foundation, the whole structure will either eventually come down for repair or cost an exponential amount of money to keep stable. If it doesn’t come down, it will be because there’s a staggering array of expensive bandages to hold it together. The latter conclusion only works until the next competitor comes along and has figured out a less costly approach.
The “Rule of 10”
In general, the “Rule of 10” states that for every phase through which a design advances, the cost for resolving issues increases by a factor of 10. Ideally, if the issue can be caught while the part is still going through initial design iterations the cost of resolution is minimized or eliminated.
Level where issue is found: The cost to resolve the issue:
If found during design phase minimal to none
If found during Prototype $X (the cost of the part itself)
If found during initial assembly $10X
If found during mass production $100X
If found during distribution $1000X
If found by the End User $10000X
By the time our customers are ready to automate they have normally advanced through the stages of design, prototyping, and initial assembly. They are now looking for solutions to automate the assembly of their products that often have inherent issues that may become visible only when scrutinized under the light of automated assembly.
Does designing for manufacturability mean changing the entire product?
Not necessarily, but that doesn’t mean it will be inexpensive to resolve. It might mean changing the thickness of the sheet metal to a common gage, or making two asymmetrical parts symmetrical, moving a hole away from an edge, or changing a soldered joint to a fastened joint, adding a series of tapers to a part so it can be injection molded instead of cut by an EDM machine, doing a basic tolerance study, or just changing a blind hole to a through hole. These changes may be minor at the part level, but when the change is propagated through adjoining components and the dominoes start to fall it may cost $100X the price of the part, or more.
Muda is a Japanese term popularized from the Toyota Production System that generally denotes a process that is required but is unproductive, wasteful and does not add value. One example might be automating a drilling process where lubricant is required. Not only is it extremely messy and requires constant maintenance, but it implies pumps, reservoirs, a filtering system, waste disposal plans – all of which add enormous cost just to drill a hole. On the other hand if a different approach can be adopted and the hole is preformed or cut by a procedure that doesn’t require lubricant, this automation process just became much more cost effective and maintenance friendly – Lean, you might say.
Why does Setpoint care as long as we get the job?
Our bottom line depends on how satisfied our customers are with the end product we develop. We want to help our customers ask the right questions during the early phases of design to help them develop a product that is automation friendly and designed for manufacturability. No muda. This allows us to deliver a solution that is robust, cost effective, and lean in every aspect.
The days of “throwing a design over the manufacturing wall” and waiting to see how they build it have come to an end. Lean automation may be expensive, but inefficient automation based on a half developed product is out of budget.