Designing for the bottom line
Bruce Rayner
EETimes Supply Network

11/01/2004 10:00 AM EST
URL: http://www.eetimessupplynetwork.com/51200541

Convincing your CFO to invest the company's money is tough. You have to show a significant need, an adequate return on investment and an implementation time line measured in weeks, not years. To boost your chances, you'd be advised to prove there's a relationship between the investment and the bottom line.

Well, here's some good news for those of you championing investments in design-process improvement. A recent survey conducted by Electronics Supply & Manufacturing and Design Chain Associates looked at the relationship between a number of formal design processes and business performance and concluded that, yes, there is a relationship.

The survey examined 11 design processes and the relationship between the number and seriousness of various design problems that companies encounter. The conclusion: The more formal design processes a company has in place-that is, the "more sophisticated" the company's design environment-the fewer and less severe the problems it encounters. The survey also uncovered a strong relationship between the time that procurement and engineering professionals spend correcting problems and the sophistication of the company's design processes. The more sophisticated the company's processes, the less time it spent chasing down problems.

The 11 design processes the survey analyzed ranged from fairly basic, like whether the company has a process for assigning new part numbers and for creating an approved vendor list (AVL), to more-complex processes like component life cycle management and commodity teams.

On the other hand, more-complex processes such as component life cycle management and commodity teams are less common. In fact, less than half of the respondents said they have these two processes in place (see chart on the top of page 18).

The survey uncovered significant patterns within the distribution of the 11 processes. Using a statistical tool called cluster analysis, three distinct groups of companies emerged. Basically, cluster analysis uncovers groups that are as internally similar as possible while at the same time are as different from each other as possible.

Eighty-one respondents out of a total of 271 were classified as Cluster I. That is, their companies are the least sophisticated, with an average of just four relatively basic design processes out of 11: AVLs, new-part assignment process, earlier supplier involvement and reuse of existing components.

Cluster II companies are somewhat more sophisticated, with an average of seven processes. These include the four Cluster I processes plus supplier approval process, nondesign engineering stakeholder involvement and component qualification process. Cluster III companies are the most sophisticated with an average of 10 out of 11 processes, adding component life cycle management, commodity teams and risk assessment process.

As might be expected, larger companies tend to be more sophisticated for the simple reason that they have been around longer and have more-complex organizations that require formal processes. Of companies with sales below $25 million, 57 percent are in Cluster I while 82 percent of companies with sales between $1 billion and $5 billion are in Cluster III (see chart, page 18).

Within the sample of 271 respondents, 35 said their companies have implemented all 11 design processes. This group is uniquely qualified to evaluate the relative impact of all 11 processes on the bottom line. This subset ranked early supplier involvement, nondesign engineering stakeholder involvement and commodity teams as having the most impact on the bottom line (see chart, to the left).

These three processes have one characteristic in common: They all require significant collaboration, be it with cross-functional teams inside the company or peer-to-peer relationships between companies. This level of collaboration is often the most difficult to formalize and manage. But according to the survey, when these processes are in place, the payoff can be significant.

Of course, the 11 design processes are not mutually exclusive-the more basic processes such as a new part assignment or AVL processes are necessary preconditions to implementing more-sophisticated processes such as commodity teams. Indeed, a company can't expect to have effective commodity teams if it has no supplier approval or component reuse process in place first.

It's intuitive that the more formal processes a company has, on average, the fewer the number and seriousness of problems. The chart at the top of the page ranks a short list of problems that respondents reported. Supplier capacity problems are at the top of the list for OEMs, while purchase order problems are the top concern for EMS providers and ODMs. A weighting of all those problems according to a measure of the bottom-line impact reveals a correlation with the three clusters (see second chart from top).

This conclusion is profound: The more design processes a company has, the fewer and less serious the problems it faces. Not having to fight so many fires frees up procurement and engineering staff to perform the jobs they were hired to do: create new products and ensure an optimal supply chain.

For survey information and how to benchmark your company's design processes, contact Bruce Rayner at brayner@cmp.com.

---