Group Actions
Engagement
  • Author Author: kas.lewis
  • Date Created: Date Created
  • Views 664 views
  • Likes 0 likes
  • Comments 1 comment
Related
Recommended

A Project Failure Due to Short Sighted Financial Considerations

kas.lewis

When I was a co-op student a number of years ago I was tasked with automating device testing for a specific company. The testing that was in need of automation was taking up valuable hours each day. One unit could easily take 2 hours a day for its first week and then twice weekly after that slowly decreasing to once per week. The manager in charge of this project had decided that using a software suite that was designed for such tasks, such as NI’s Labview was too expensive and that we could do the same thing in C#. The time allocated for the project was two months (I was on a four month co-op term).

 

2494022388_021f373229.jpg

 

Included in the automation system was the integration of a photon counter, network connected multimeter for voltage and current measurements as well as controlling the device under test to load different software between each test (there were other modules that I either don’t remember or can’t explain as it would be to company specific). The system was then supposed to log everything in a database and track performance of each device over time. While these goals were definitely reachable as I had seen on a previous co-op term at another company, that was all achieved with the help of LabView.

 

After innocently asking at a few meeting and close to the third month of my co-op term why we were going the C# route and not the LabView route I was told it was to save money and it would not be to much more work to complete in C#.

 

I left after my 4 months but kept up with two of my coworkers who had graduated from the same university as myself. It was interesting to find out that what code had been written while I was there (multimeter control, software loading and maybe a few other smaller modules) were being run by a python script, and the grand test software package was never completed as requested. I’m not sure how much time they still spend on testing but I’m sure that had they settled for something like LabView in the beginning the time saved and overall efficiency would have been much higher, and the overall money saved would have covered the cost of the suite they selected.



Have any of your engineering projects ever failed due to factors beyond your control? Share your story here!





Tweet

Follow @Kazzzzzzzzzzz

Follow Kas' blog

  • Hi Kas,

     

    Once again, I have seen a lot of this type of mistake in companies large and small.

    Everyone assumes they can do it quicker, better, cheaper than the project could be done by using off the shelf products to speed up the build.

    I once had Quality Control engineer decided he could build the best automatic check software rather than buy to top of the line off the shelf product.

    I finally found out about it after he had spent over $25,000 USD on a tool I could have purchased for $5,000 USD.  If I could I would have fired him, but my boss would not let me.

     

    So I recommend that in the future, always do a cost benefit analysis to look at available products verses home built.  Always double the cost/time/effort estimated by the we can build here crowd and you will find that you can often buy cheaper/faster/reliable than the build it crowd.  Oh and always ask them how they are going to test and verify the home built unit, what the time for rework will be and are they willing to work free overtime to keep it online at the expected production rate.  That usually ends the argument.

     

    DAB