It’s early in the boating season. It’s a beautiful Saturday and wee’re wakeboarding, My wife is driving. I am getting ready to line up to jump the wake and all of a sudden she cuts the throttle and then guns it again. I just let go of the rope and wait for her to come around so I can find out if it was the dirty dishes left on the couch or beard shavings carelessly sprinkled on her face soap. She said the boat just stuttered without her touching the throttle. Hmmm really? As we are talking the boat just stalls. Ughh!, and we are not close to the house.
The Apex Ridge article series covers a diverse set of topics that relate to many of our reader’s work, interests, and experiences. The articles are inspired by industry experiences with the intent of sharing, educating and assisting you with your career challenges and growth. The content is targeted for a diverse audience with backgrounds even extending beyond engineering (Hmm talking to you project and business managers). My hope is that these topics inspire you to have discussions with your colleagues or right in the comments of the series. I look forward to seeing you on-line soon.
Some leaders mistake “customer focus” to mean they have to serve all of the customers needs or respond to every request from the field. There are multiple needs I have for a vehicle. Because of this I have more than one vehicle. Many manufacturers have tried to make a car or truck that does everything. It often just ends up being a vehicle that is great at nothing. (See Pontiac Aztek)
Know your target. Make goals, make compromises. Don’t commit to high reliability without making sacrifices. Something has to move either it be schedule or new technology development. There are no worse words than a leader saying “…and it must be highly reliable” without discussing the cost of pursuing that reliability goal. Know up front if you are willing to trade reliability for growth of technology or time to market.
Stress Margin is an interesting topic because our gut reaction is “more is better.” But more isn’t better. The key is figuring out “How much and where?” this is where the attention should be paid.
Too much stress margin can end the project the same way making material too thick will turn a plane into nothing more than a crappy diner with too much security. It’s the correct margin that is needed. How do we select the correct margin?
When a product has broad usage profiles how do you create one for a test protocol. Do you select the easiest? Obviously not because there is risk of a high failure rate from poor evaluation. Do you select the hardest use-case? You could but there is the risk you spend too much time and money making the product overly robust for most of your customers.
In these cases where test time is limited and only a single use case can be tested I create a “Composite” use-case. It captures many of the different types of stresses from the full list of use-cases but keeps an accumulated stress on the design that represents at least 98% of the customers.
A common progression (stages of awakening) for an organization that has a minimal application of reliability tools to one that is large market holder with the right reliable product is often like this
Stage 1) “Reliability” testing is mostly re-labeled verification and validation testing. It measures if the design does what it is supposed to do at the end of the program. Tests are executed with the intent of passing, not learning about the product or it’s performance under variability. The organization experiences a large field failure surge due a single or multiple issues. The pain of this experience in dollars, market image, and lost resource through “recovery phase” awakens them to the high ROI of incorporating reliability tools early and throughout the product development process
Many product development teams use a contract manufacturer (CM) to develop and manufacture their product. It’s the “develop” term that has limitations that may be unforeseen when engaging. Many companies use the CM to assist or even eventually entirely execute the product development. This can have great results or be problematic. But let’s talk about the arrangements that work well. Troublesome CM’s are a topic for a different time.
In many program cases I see teams “testing to pass” when they should be “testing to improve”. Testing to pass is putting your best foot forward. There is a “mark” and you are going to hit it so you can advance to the next stage. Testing to Improve is looking for defects and response to inputs. The motivation for each is very different. The risk for leaders is overseeing teams who view their role objectives to be in line with “testing to pass.”
Iideytinnfg pantrtes in dtaa is why we pforrem stsiattaicl aalisnys. It is not an ecaxt sicncee but iarmitofonn can be exeaaltpotrd eevn if the dtaa is nsoiy or inmptcolee.
The Rseaon Wuleibl asaynlis is ieeeptmnlmd as a tool is to look for pntertas and tehn cetare a charisotairceatn of a cmomon boeihvar. Taht charcoeatistairn can tehn be uesd to unsretadnd uocnpimg silimar pntarets. The raoesn you hvae been albe to raed tihs alirtce is beusace yuor biran fgriued out taht eevn tuhogh teh wodrs are samerbcld the frsit and lsat leettr and the wrod lgnteh areare corcert.
Uisng tihs prttean and the sceentne coxnett and rninaemig ltrtees and cmipanrog taht asnaigt yuor vbualrcaoy you can qulikcy dremtinee the menanig of the sectnnee.
It’s spimly a mteatr of aniaslys, pteratn chractrsaoietian, and tehn precdoitin by aiplnypg taht chariocatestrian to a new dtaa set.
Wlel Dnoe !
The challenges of keeping reliability at the resource and schedule negotiating table mid-program can be greatly helped by a solid effort to connect it to the business case when the product program is created.
It’s the ambiguity of how reliability makes the company money that makes the case in live negotiation so difficult. Everyone knows that reliability affects sales, marketing, warranty expense, future development. But how do you compare that to the urgency of time to market or cost point? If you, as the reliability representative, can point to a specific quantitative connection made to the business case for the product you have one.
So I wrote about how Mercedes has crummy reliability per Consumer Reports. I then threw a zinger at the end of the article saying I do have a Mercedes in the stable. So here is moment in time with her before we put her away for the winter last year. You will see the love/hate bloom and a benchmark with a 40 year old Japanese car. [Read more…]
Leading reliability efforts is a hard road. The resource required to improve reliability is significant. The resource to demonstrate reliability is gargantuan.
There are a few ways to demonstrate statistical confidence in reliability of a design. Each has it reasons for selection. here are three of my favorites [Read more…]
I recently came across this announcement that HP is going to develop computer systems and peripheral devices that are optimized for environments where they are cleaned frequently, often medical environments. In the medical and scientific fields cleaning solutions are a significant stress for plastics, inks and touch control surfaces. It’s a smart strategy to evaluate where your products are used and optimize your test use cases. It most often will lead to more accurate test results and reliability projections. It may, as well, lead to a market opportunity that was previously unknown, and if you are the first among your competitors to acknowledge this need and create a specialized or derivative product you now grabbed a portion of the market that you may have not previously held.
A few others
- laptops that are impact and dirt resistant
- Kids winter coats with glove attachments
- Sport tuned models of economy cars
- Stackable chairs
- Folding bikes
- Standing desks
I was asked “Do you know about Singularity design?” I hadn’t heard the term in that context before.
It’s the concept of not approaching design from multiple disciplines. The design process is done with an approach and knowledge base of all needed disciplines at once and in conjunction. The electrical system isn’t designed by an electrical team and the mechanical by a mechanical team. The “Design Team” designs both simultaneously. Team members knowledge might be rooted in one discipline but there is clearly no boundary to their knowledge of other disciplines. This would be a mechanical engineer who has designed a PC board before.