I’ve been loosely following Boeing’s 787 Dreamliner project and thought I’d throw out a few high level lessons learned. Here are two interesting articles on the project, one written in June 2006 and the other in June 2009.
Let me start off with the bottom line – the project is delayed over 2 years and is costing Boeing dearly in terms of money and reputation. What happened?
- Boeing made a HUGE gamble on new technology. They didn’t come up with the name Dreamliner by accident. IF it can live up to its marketing literature, Boeing will dispel notions that its glory days are over. New technology poses complex risks to project estimation. In the June 2006 article, it was telling that engineers were already shouting at each other during meetings. We can assume that this means that marketing and the stock market drove the original delivery date, not good project management. If you are using new technology, expect things to take longer. Sometimes a lot longer.
- Outsourcing. Boeing made a decision to outsource over 50% of the work. This made their project success dependent on their vendors. Vendors, if capable, can deliver great products and services. However, regardless of how well the vendors were selected and monitored, the project was doomed because of changing requirements. Because the technology was new, requirements were changing all the time, and vendors couldn’t respond in a timely manner. What is troublesome, is that if requirements were ill defined, how do we know that they’ve been completely flushed out. Just recently, Boeing determined that extra reinforcement needed to be applied to where the wings join the fuselage. Hmmm – are there any more surprises forthcoming?
- Just in time testing – or… I hope this works or the schedule gets extended. There is a lot of emphasis lately on Rapid Application Development in the software development world. There seems to be a push in iterative development in many industries, and Boeing is no exception. This works well on projects that have been done before, but when using new technology it poses significant problems in meeting deadlines. Boeing should be hoping that nothing new is discovered as they enter their flight testing. Yeah, right. New technology, multiple vendors, poor requirements, and there won’t be a significant problem during the next phase of testing? I don’t think so. Why? Because I haven’t seen anything to convince me (although to be fair I’ve never been in one of their project management meetings) that they are changing the way they do their estimates. There is a LOT of pressure on Boeing to get the Dreamliner into production. This is driving their estimates, not sound project management principles. How do I know this? They keep missing deadlines. Hello…… have they learned anything?
So… what are a few take aways?
- If using new technology, ensure you buffer your estimates to allow for setbacks
- Don’t let marketing drive your schedule
- You need good requirements BEFORE you develop your estimates
- If you utilize vendors, allow for time to coordinate changes and setbacks
- If you want to do rapid development/production, you need good requirements
- If you continually miss your estimates, find out why and adjust your process accordingly.
All the best…
All the time…
Related articles by Zemanta
- Boeing 787 steps closer to first flight (seattlepi.com)
- Update on Boeing 787: Intermediate gauntlet testing underway (seattlepi.com)
- Supply chain concerns Boeing (seattletimes.nwsource.com)
- Months of delay likely for 787, Boeing partner says (seattletimes.nwsource.com)
- Boeing 787 Dreamliner engine test with smoke (video) (seattlepi.com)
- Qantas puts yet another dent in the future of the Boeing Dreamliner (gadling.com)
- Why does Boeing think it can finish the 787 on time? (seattlepi.com)
- When (Outsourced) Projects head South (brandimpact.wordpress.com)
- First flight of the Boeing 787: T minus (at most) 11 days (seattlepi.com)