The Situation – Expensive Manufacturing
A major US medical device company came to us with a problem. As one of their new products gained market share, it became clear that production costs were too high. One particularly laborious part of the production process involved very precise sewing, which takes 15-20 hours per unit assembly. This rate of throughput presents an obstacle to scale the product line. If the assembly process — or the sewing process in particular — could be accelerated, then scaling up the product line may be simplified. For this technical challenge, the client recognized the need for outside expertise and engaged SGW Designworks for their diverse background in developing manufacturing processes in several industries.
Development Approach – Proof of Concept
The project began with a review of goals, constraints and expectations of all teams involved. These initial sessions were held at SGW’s office in Boise, Idaho, with one client team member onsite and others teleconferencing in. With clear alignment on project goals, SGW’s team staged collaborative brainstorming sessions to identify potential paths that could have a positive impact on assembly time of the product. Next, fast cycles of “Build-Test-Learn” were used to validate or invalidate the conceptual approaches in parallel. These cycles included design and analysis work in 3D CAD and fast development of 3D printed and machined subassemblies. Based on the results of the first cycles, the concepts were ranked in terms of viability and potential impact. Results and recommendations were reviewed with the client to ensure that development work was still in line with their goals.
Prior to starting the project, there was uncertainty as to whether a viable solution to the problem existed. Initial findings indicated that three possible solution paths could likely result in improved production times with fairly low technical risk. The most attractive approach was refined, and SGW developed a proof-of-concept cell with a small robotic arm and custom fixturing and programming to validate the approach. The cell was able to bring in material for the assembly / sewing process, perform the sewing operation, move the partially finished unit to an automated inspection tool, and bring in the next unit for sewing.
Some unexpected challenges were that fixturing, clamping and positioning were critical parts of the solution (as opposed to sewing). Humans do this well, and SGW needed to mimic the human approach, which became areas of parallel research. Solutions for clamping were developed during the course of the project, and the client adopted these solutions for production even before the project was completed. Throughout development, SGW and the client were in constant contact, ensuring that changes in direction were decided upon mutually and that work aligned with the client’s goals.
The project took three months from kickoff to wrap-up, resulting in a proof-of concept system demonstrating a viable solution to the problem of increasing production throughput.