Penny wise, pound foolish

DN Staff

November 15, 1999

10 Min Read
Penny wise, pound foolish

Looking to cut costs, an earthmoving equipment manufacturer in South Africa purchased a screw for the axle assembly from a supplier with the lowest price. No big deal, or so the firm thought. But shortly after the equipment went into production, the screws began to fail in the field in a very unusual way. Instead of breaking or shearing at the head, which is a more typical mode of failure, they were splitting down the middle.

A failure analysis later revealed that the problem was two-fold: First, the screws had been insufficiently heat treated after electroplating, causing hydrogen embrittlement; and, second, there was a basic material flaw running right through the centerline of the failed parts. "You couldn't see these problems with the naked eye, but those screws were just like a time bomb waiting to go off," is how Harry Lyons, a product manager with Danaher Tool Group Europe, describes the situation.

Ultimately, and at enormous cost, the company had to recall parts from Africa, the U.S, and Europe. They wound up replacing the faulty hardware with a more costly but higher quality screw from the Scotland-based Danaher Corp. End of quality problems.

Achieving high quality and reducing cost do not have to be mutually exclusive goals. Better quality components and parts may cost more money in the short run, but can actually save companies money in the long run.

Net present value. As was the case here, most organizations are not fully aware of the potential consequences of using poor quality parts until it's way too late. If only design engineers had a way of knowing how much cheap parts may wind up costing in the long run, they could help discourage others in the company who actually purchase the components from being tempted by them in the first place!

Actually they do. It's called net present value (NPV), which is a formula for calculating a project's net worth or return on investment (ROI). Many design engineers learn about NPV in an industrial engineering course and later apply it on the job to justify projects to upper management. However, most of them probably do not recognize that the formula is a useful way to examine the financial impact that poor quality components can wreak on a design project.

NPV takes into account the time value of money--namely that a dollar today is worth more than a dollar tomorrow--by discounting expected cash flows (both expenses and revenues) from each year of a project (see baseline calculation below). These cash flows are then summed to produce an NPV for the project.

The payback period can also be determined from the NPV calculation. It is the amount of time it takes for the NPV to equal zero, which is the point at which the cumulative forecasted cash flow equals the original investment. A project that produces a positive NPV is generally considered to be an attractive investment opportunity, as is one that has a quick payback.

Net present value: -C0 + (sigma) CN/(1 + r)n = $151 millionInitial investment = C0 = $100 millionSales revenue = CN = $75 millionDiscount rate = r = 15%Life of project = 5 yearsPayback period: Within 1.5 years

While the formula is relatively simple to crank through any good financial calculator can do it for you--coming up with the specific numbers to plug into the formula is another issue entirely. Cash flows--which are the dollars going out and coming in for a project--are usually based on "best-guess" estimates from engineering and sales. Most companies assume a discount rate of between 10 and 20%.

However, in nearly all cases these numbers are predicated on a best-case scenario. In other words, the people responsible for pulling them together assume that things will work out according to plan. Even when buying cheap, substandard components, hardly anyone--as evidenced by the people working for the South African equipment manufacturer--expects anything to actually go wrong.

But when it comes to poor quality, things can and do go wrong, all the time. As the charts on the preceding page illustrate, substandard parts or components can impact the NPV of a design project in a variety of ways. Time-to-market may be delayed as the design team scrambles to correct unanticipated quality problems. An extra infusion of cash in the project will probably be necessary in order to pay for extra engineering resources or overtime. Should quality problems arise only after the product hits the market, sales could plummet far below initial revenue projections. And that's not even including repair, replacement, and warranty costs. Any one of these problems can totally change the economic picture of a project--for the worse.

A tough sell? Nonetheless, quality remains difficult to pitch. "If companies would stop focusing on the initial purchase price and look at total system costs, quality parts and components would not be such a tough sell. But it's often a challenge to convince someone to spend more money on a higher-quality part, especially when the physical differences aren't obvious," says Danaher's Harry Lyons. Although his company's line of Holo-Krome fasteners has a unit cost that is 40% more than that of the competition, he likes to tell engineers that they are not in fact the world's most expensive fasteners.

"The most costly part is not always the one with the biggest price tag. It's the one that fails because of an incorrect grade of material, the wrong hardness, a poor fillet radius, or other quality issues," says Lyons.

Price aside, the fact that nearly every supplier today is touting that they have the best quality product puts engineers at a disadvantage. And though they might not be the ones making the actual purchasing decision, they need to make sure that others in the organization who are don't buy poor quality.

"The engineer's task, dilemma, and challenge is to be able to differentiate between suppliers," says George Jaffe, executive vice president of Schneeberger Inc., a manufacturer of linear bearings. Schneeberger is committed to quality and engineering education, as evidenced by its sponsorship of the annual Design News Quality Award. Each year, the award recognizes one company that best exemplifies quality in its approach to product design and development. Past winners include Xerox for its revolutionary new digital copier technology, and Cincinnati Milacron for its line of innovative centerless grinding machines.

"The process should be expanded beyond traditional product features, functions, and benefits to include company reputation, tech support, warranty, price, protection, and so on," says Jaffe. "Engineers should also ask how the product has been tested, whether test data is available, and user references."

Last but not least, engineers should always remember that when a deal seems too good to be true, it probably is.

Quality affects Net Present Value

Sales may be lower than expected The steepness of this line indicates the relatively significant impact that sales estimates make on a project's ROI. Quality problems can affect sales drastically, putting a big dent in the NPV.

An extra infusion of cash may beneeded to fix quality problems Delays and quality problems almost always cost money, and these are typically dollars that are not factored into the original NPV equation. These unplanned expenses put projects deeper in the hole.

Time-to-market may be delayed Project delays-whether related to quality problems or not-are expensive because they push expected sales revenues further into the future. In addition to reducing the project's NPV, even a small delay can add a significant amount of time to the payback period.

Although they may save some money upfront, low-quality components and parts can wind up costing a lot more money in the long run. Delays, cost over-runs, and lower-than-expected sales due to quality problems can totally change the economic picture of a project, while the slight added cost of a better component typically has only a minor impact on the overall NPV.

Tips for checking quality

What steps can you take to ensure the quality of parts and components that go into your design? Here is advice from four vendors of different products.

Steven Verrett, product engineer, Omron, manufacturer of switches, relays, and other products.

Check process-quality control data.

  • Tour vendors' factories. Many large manufacturers do that, particularly when purchasing large volumes of product.

  • Check whether there is 100% outgoing inspection. Most relay manufacturers have programs for that.

  • Learn the vendor's record for service and support. Those factors can add to your comfort level with the vendor.

  • Purchase as standard a product as possible.

  • Make sure the product won't be discontinued soon.

  • Build a relationship with your local sales representative.

John O'Brien, director of global sales, Penn Engineering & Manufacturing Company, fastener manufacturer.

Analyze design integrity. For example, check if thread strength of fasteners matches the thread strength of the mating part, and whether the amount of raw material used is the optimum for performance, weight, and strength.

  • Inspect dimensional tolerances. For example, look for variations in a fastener's threads.

  • Check thread fit and see if equivalents all meet the same specifications--and then test them.

  • Determine prevailing torque as specified by appropriate standards such as MIL-DTL-25027, which sets minimum and maximum breakaway torques and the number of times the fastener should resist turning.

  • Examine heat treatment and plating. Poor heat treatment can cause a fastener to fail.

James R. Kimzey, executive vice president Research and Engineering, Baldor Electric Co., motor and drive manufacturer.

Design for simplicity of manufacture, which lessens the chances of mistakes being made on the factory floor and lowers the time and cost needed to produce the most value for the customer.

  • Follow your designs to the factory floor to help get them into production and tweak them for manufacturability.

  • Work closely with the purchasing function and remember that superb documentation does not automatically guarantee that purchased components will perform as envisioned in the original design. Eyeball-to-eyeball communication is also needed.

  • Remember that the absolute peak of quality may not necessarily be the best choice in some applications--cost and time-to-market must be kept in mind with the best value being the desired result.

Kurt Boegli, chief standards engineer, Phoenix Contact, manufacturer of electrical/electronic connecting systems and controllers.

Choose the supplier that is best qualified to answer all technical, quality, and product approval questions.

  • Select products and components that have been in use for awhile and have accurate reliability data.

Sign up for the Design News Daily newsletter.

You May Also Like