Author: Henri Azibert

An Engineer’s Point of View

The life of an engineer is quite bit more complex than most realize.  The typical image of an unassuming, small stature, thick glasses wearing individual, prominently carrying various calculating or scribing tools, comes to mind.  In many respects, however, engineers are the ones that rule our lives.  From the alarm that wakes us up in the morning to the vehicle that transports us to our destinations, to the incessant interruptions of various communicating devices that rely on satellites in low earth orbits, all these conveniences or nuisances of modern life and all the machinery required to manufacture all the pieces were dreamed of and then designed by engineers.  Yet, engineers are often criticized for being “risk adverse”.

Being an engineer is quite difficult.  A fundamental reason is that engineers must deal with reality.  Unlike marketers, managers, or politicians, who can and do argue the obvious, the device either works, or it does not.  There is no ability to dispute the functioning of the mechanism.  In the manufacturing world, the pump moves the required amount of fluid from one tank to another, or it does not.  The seal contains the product or it leaks. The valve opens or remains stuck.  And when it does not do what it is supposed to do?  We ask the engineer why, and what she will do about it.

In the world of industrial engineering, machines are designed, manufactured, and sold for a particular purpose.  Designing for a function, no matter how complicated, turns out to be the easy part.  Difficulties start to creep in quickly; the first being to very precisely describe what a part should be like.  There is the specification of the material, all the dimensions and tolerances; the part must be depicted in excruciating detail.  A typical engineering drawing can easily contain over 1000 symbols.  Just two digits being reversed can lead to parts loosening up or not being able to fit.

Then, the manufacturing group wants to relax tolerances; the accountants insist on lower costs and the removal of the most important features; logistics wants to eliminate parts to reduce inventories.  Requirements routinely drift from the original goals.  Pressure, flow, and temperature capabilities are increased several times during the development.  The schedule requires 10 months if absolutely everything goes as planned, but Marketing dictates product delivery in half the time: 5 months, tops!

But let’s assume that the project went well – the design was thorough, the checking was intensive, and the testing (even though limited by the prototype budget that was slashed twice) was positive.  All that is within the realm of engineering was done to perfection.  You pulled all stops, worked evenings and on weekends, used all the shortcuts and got the project done in 6 months.

What could possibly go wrong? 

To start with, you are one month late.

Remarkably, of the hundreds of problems that could have happened, only one surfaces. Procurement saved $10K on an order by going to a new vendor.  That made the bonus for the purchasing agent.  The parts come in three weeks late and do not meet the print.

If you reject them, there will be another month delay, with extra cost for a rush order.  The purchasing agent is now a life time enemy if you make him loose his bonus. Marketing is up in arms for the delay. Accounting notes you did not meet the cost target and thus half of your 0.003% annual increase is gone.  Projected sales shortfall becomes your fault, which takes care of the rest of your merit raise.

What is the alternative?  The decision is a difficult one, trying to estimate how much of a problem the nonconformance could cause.  Under the most probable conditions, there should be no impact.  Only under negative pressure, unusually high speed, and widely fluctuating temperature will the device underperform.  This combination of conditions would be extremely abnormal and a change from the usual installation process should obviate the detrimental effects.

Do you take the risk?

With pressure coming from all sides, you decide to accept some of the parts.  A deviation is allowed.  A full inspection to find the parts that are deviating by no more than 300% from the tolerance range will be sorted and accepted.  The quality inspector needs to stay late to go through the inventory and misses his daughter’s third grade final dance recital.  He vows to make your life as complicated as possible, requiring that all dimensions be reviewed or changed, since obviously, you can use parts outside the tolerance range you originally specified.

Units are shipped and preliminary reports are encouraging.  But then, the number one salesperson decides to use the new product in the one account where he was going to be tossed out due to a series of failures relating to the fact that he had oversold the most expensive, but totally inappropriate, model in the line.  Sure enough, that unusual combination of conditions is there.  To make things worse, the installation instructions were not reviewed and the wrong sequence of steps was followed.  The device fails.  Process product is spilled, production is stopped, downtime costs pile up!  Thankfully no one was hurt, but the account is lost.  The now number three salesperson vows to get you fired for specifying such an obtuse installation procedure.

What could go right?

Everything goes superlatively well.  Sales volume is up, customers’ problems are solved, and the product is recognized in an Industry Award.  The Marketing VP gets to accept the award in a glamorous and exotic location, and he brings you back a handsome plastic recognition plaque, the size of two business cards, that gets to be prominently displayed at the Engineering department’s entrance.

The risk to reward balance is what makes engineers to be risk adverse.