Well, our new Prototyping Designline is certainly attracting some interest. Just about everyone who has called or emailed me recently has dropped it into the conversation. One rather thought-provoking example came from Leo Sennott, who wrote as follows:
Hi Max, I recently discovered your new EE Times Prototyping Designline, which caught my interest. During my 30 years of engineering in the semiconductor industry, I have been fortunate enough to see all facets of this global business through designing and developing many custom applications supporting assembly test and manufacturing requirements.
As engineers in this space, we all feel the daily demands of this never-ending changing market. In order to remain competitive, new products are required in rapid fashion with customers determining market timing. This, in itself, is a major undertaking; however, we also have a few business measurement acronyms to consider during these endeavors DFM, DFT, OEE, KPI, ASP, GM, etc...
From my experience, prototyping is an area that is most critical for new product development, but one that tends to be neglected when it comes to capital, automation, processing, and -- most importantly -- knowledge sharing.
I would be willing to bet that many prototype lines today are subject to missed market opportunities because they are still running with some archaic legacy application or numerous "Master" spreadsheets.
Some years ago, I hired a key addition to my engineering development team. Although not an engineer, this individual was proficient in web-based applications and database design, and could provide interfaces for business enterprise integration.
This web-based technology allowed for new engineering design concepts, where new levels of intelligence, management, and visibility could not only be incorporated, but could also be dynamic.
A few years ago, I was approached by a senior staff member and asked if it was possible to develop a portal specifically for managing several prototype lines. At that time, these lines were underutilized and had limited procedures with constant design rule violations. Needless to say, every build was a priority.
After several meetings of working directly in the prototype lines, the goals, and objectives of the application, were defined as follows:
Goals: Design a solution that directly interfaces the design engineering community with all prototype centers. The portal should promote, sustain, and notify a collaborative assembly development environment that provides real time, historical, and assembly data with a genealogy trace to the material sets used.
Objectives: Web portal provides a collaborative set of tools to organize and record all prototype build specifications, data collects, and manufacturing metrics.
This involved designing a custom database that specifically addresses prototype and packaging assembly needs as follows:
Clean interface with input modules designed specifically for multi-chip module assembly.
Individual modules by process operation to improve clarity, accuracy, and cycle time (die attach, wire bond, mold, laser mark, singulation).
Transfers to global prototype lines.
Administration rights allowing for complete portal management.
Real-time data processing for build status, alerts, events, action items, and hold conditions.
Consolidated manufacturing data collects -- prototype performance indicators.
Transform prototype data collections into an active knowledge framework with search engines allowing for historical data access with "drill down" capability providing greater analysis.
Capture all build information (specifications, wafers, BOMs, assembly data results, material genealogy images).
The results were as follows:
General: Paperless, searchable, elimination of emails and phone, live status, e-notification system.
Overall efficiency: Parallel builds, increased number of builds, reductions in cycle times, hold times, and approval times.
Knowledgebase established: Complete genealogy of BOM's, die placements, wire-bond lengths, loop heights, bond line thickness, RF tuning, PCB's, saw accuracy, etc.
Design user forum established.
Real-time key performance indicators: Builds (per schedule, product line, product, design engineer, customer) ; Equipment metrics: number of die placements, number of wire-bonds, etc.
Well, now I'm left wanting to know more. Perhaps we'll be able to persuade Leo to come back and elaborate on some of the topics he introduces above. In the meantime, what are your experiences in this area?
Hello all and thanks Max for replying to the comments.
These are very good comments and I can see many questioning if this business function is really "most critical". I would probably agree that the comment might NOT be a "most critical" business function across all industries.
However, my background is the RF industry and I do believe that "most critical" is an accurate statement.
This space over the last several years can be compared to the personal computer evolution. The smart phone / tablet market has changed so rapidly that a single missed opportunity can sink your stock in a day.
Why I still hold this opinion is that these products, as well as market timing, is determined by the likes of Samsung, Nokia, Apple, Huawei, et al. The performance specifications, package type and footprint are extremely difficult to achieve and typically will require several design spins. All eyes and ears are waiting for new results out of these prototype lines and ready to act.
@Tom: I do have some doubts that any part of a product development process can be labeled the most critical stage...
Well, I can't argue with you there -- every stage is critical as far as that goes -- but remember that it's not I who said it was the "most critical" -- it was the guy who wrote to me -- Leo Sennott -- so I think we will have to bring him into the converstaion ... I shall "ping" him now...
Not so bitter, Max! And i should know better than to raise the question at a site populated by electrical designers, but I do have some doubts that any part of a product development process can be labeled the most critical stage. Sounds like a good poll question for the community!
I do agree that prototyping is critical, whether one is working on hardware, software or both. Wearing my software developer hat, one of the first things I typically did was to prototype the GUI to get early feedback from propsective users (there are many tools available or this, eg., Balsamic).
These days there are plenty of options for going beyond bread-boarding to get a product in hand that almost looks like the final one! 3D printers (stereolithography prototyping machines for those you more technical!) have enabled the modern day startups that can produce almost functioning & enviable proto's.
The most value I get out of protos are feedbacks on problems in the design that often escape brainstorming on a white board.
Max: You gave us a lot to chew on there, but I'm not sure I like the taste. True, the economics of a faster supply chain, fast-evolving technology, fickle consumer tastes and more contribute to the design of new products, and I'm not sure prototyping is THE most important factor anymore. It's important, don't get me wrong. I believe in prototypes, QA, analysis, focus groups and all the conventional steps in designing products. Isn't it more like a step in staircase now, rather than the most important step?
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.