Group Actions
Engagement
  • Author Author: kas.lewis
  • Date Created: Date Created
  • Views 2367 views
  • Likes 5 likes
  • Comments 7 comments
Related
Recommended

A Project Failure Due to Lack of Planning and Time Management

kas.lewis

I was recently working on a project where the goal was to control a rocket's altitude and recovery system. The basic goal was to have an altimeter to know at what altitude the rocket was at in order to deploy the drag chute and then at an acceptable speed deploy the main chute. For a first time attempt at designing a system to work for this application this goal would have been a good first step however, the team leader decided to go big or go home.

 

rc_top.jpg

 

With just five months till launch date new sensors were still being added to the already over loaded system (there were approximately 10 sensors crammed on a BBB custom cape) and no board had yet to be spun. The software was also very badly defined as the goal of the project had become clouded with big dreams of short term impossibilities.

 

With just two months to go and no board spun or any hardware tested, we were still trying to add a remote ignition to the rocket, figure out how to power the Beaglebone Black from the single power source while also having power for igniters, long range transmitters and a GPS system. With some things finally nailed down a rev 0 board was made with much difficulty (letting a sponsor choose your fab house can really slow things down), but ordering components took another week just to fill a cart.

 

With no real end goal insight and a lack of ability to populate the boards, most members of the team disappeared and it was left to one overly optimistic soul to attempt to save the project. With a week left to launch date I got a early morning email saying that poor soul had tossed everything out as it was not possible to get anything meaningful working at the last possible moment.

 

Sure the team lead had started with good intentions, but his inability to nail things down in a timely manner as well as to have a solid plan how to move forward cost the team their hard work and custom electronics they so badly wanted. Hopefully moving forward this lesson will be acknowledged and learned from so as future projects will be more attainable and structured, only time can tell.

 

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





Tweet

Follow @Kazzzzzzzzzzz

Follow Kas' blog
Parents

  • Hi Kas,

     

    I have been on several projects like this and it taught me that someone needs to check and verify every part of the project and establish firm timelines for subcomponent submissions.

    Everyone assumes that everyone else will do their part.  After 30 years on very small and very large projects I stick with the motto "Trust, but verify!"

    Every project needs someone who is willing to act as the adult supervision.  Its a thankless job, but essential.

     

    Have you considered establishing a testing simulation?  We found that you could quickly work out timing issues and interfaces early and then link in the actual Hw/Sw as they became available.

     

    I found that approach worked well when all of the contributors could not be on site all of the time.  It also let each group to do quick prototypes to resolve issues.

     

    All I can say is that given your current issues, I suspect you will take a significant time delay before you get everything ready.

     

    Try holding a phone conference so that everyone can identify any ideas on how to get things back on track.  If you do a reset of your timeline now, you might get everyone's concurrence to make the deadlines.

     

    Good luck,

    DAB

Comment

  • Hi Kas,

     

    I have been on several projects like this and it taught me that someone needs to check and verify every part of the project and establish firm timelines for subcomponent submissions.

    Everyone assumes that everyone else will do their part.  After 30 years on very small and very large projects I stick with the motto "Trust, but verify!"

    Every project needs someone who is willing to act as the adult supervision.  Its a thankless job, but essential.

     

    Have you considered establishing a testing simulation?  We found that you could quickly work out timing issues and interfaces early and then link in the actual Hw/Sw as they became available.

     

    I found that approach worked well when all of the contributors could not be on site all of the time.  It also let each group to do quick prototypes to resolve issues.

     

    All I can say is that given your current issues, I suspect you will take a significant time delay before you get everything ready.

     

    Try holding a phone conference so that everyone can identify any ideas on how to get things back on track.  If you do a reset of your timeline now, you might get everyone's concurrence to make the deadlines.

     

    Good luck,

    DAB

Children
  • in reply to DAB

    Hello DAB,

     

    There were some suggestions of incremental building and testing. A lot of the components we were using can be bought in modules, so at the very least the software could have been tested with each hw module individually. Unfortunately the team lead had his own ideas and on top of that he was always deferring his final answer until talking to someone else who was not directly involve din the project, did not understand the team members time constraints or level of skill. As for meetings, we had weekly meeting, unfortunately a lot of time was spent by the team lead moving around finding a quiet spot,  a spot where others would not be disturbed or where the internet connection was good enough for us to hear what he was saying.This is a point I think that is lost on a lot of people, bad meeting don't produce no result they do worse than that, they can demoralize members and end up causing worse results as people feel the project is without direction and time is not important as it is being wasted by management in unfruitful meetings.

     

    Thankfully there is a new team lead who is very much more interested in keeping it simple and doing incremental build that add small features at each iteration. It will cost a lot more but in the end it will have a 100% chance of working better than the previous philosophy. Only time can tell how it works out this time, I'm hoping that it goes well.

     

    Kas