Breaking News
Blog

Facts and data versus heuristics and hope

Ron Collett
11/15/2010 03:45 AM EST

 6 comments   post a comment
NO RATINGS
View Comments: Newest First | Oldest First | Threaded View
krisi
User Rank
CEO
re: Facts and data versus heuristics and hope
krisi   11/30/2010 12:40:24 AM
NO RATINGS
thank you Ron, agreed...pls pass the info on to executives ;-)

RCollett
User Rank
Rookie
re: Facts and data versus heuristics and hope
RCollett   11/29/2010 11:54:18 PM
NO RATINGS
Kris, My observation is that giving a team an unrealistic schedule is usually counter-productive b/c it is a demotivator. Far better is to give a team a very aggressive -- but achievable -- schedule. When I say "very aggressive," I mean a schedule that assumes the development team's prodcutivity will be best-in-class. In other words, in order to achieve the target schedule, the team must perform at or near best-in-class. The result (based on tracking this phenomenon on hundreds of projects across the industry) is that the productivity of those teams are far above norm -- and the projects are are on schedule (slip is less than 10%) and come in within tolerable budget margins. Thanks for your comment. Rgds, Ron

RCollett
User Rank
Rookie
re: Facts and data versus heuristics and hope
RCollett   11/29/2010 11:44:51 PM
NO RATINGS
@Robotics Developer, There are definitely companies making appropriate up-front tradeoffs -- Intel for example. Here is an abstract of a paper presented by Intel at an industry conference held earlier this year devoted to project planning strategies: "Effective planning for complex design projects requires understanding the tradeoffs between resources, complexity, schedule, and risk. Resource decisions are based on a particular project's priorities, which may impact team productivity [because team size might need to be increased to hit target schedules, and increased team size will reduce productivity -- but increase throughput], and hence are important to comprehend in any systematic approach to benchmarking or planning. An empierial model for silicon design is built from available data: both from internal and external sources. This mdoel can be used to understand reslationships between team size and productivity, complexity and duration, risk and performance to schedule. Use of the model to predict organizational limits for complexity is discussed along with how key trends like reuse and modularity are unavoidable responses to current trends. This model is used throughout Intel to help make silicon design resourcing decisions . . . ." In light of the above, perhaps it's not surprising that Intel's financial performance continues to be exceptionally strong. Rgds, Ron

Robotics Developer
User Rank
Rookie
re: Facts and data versus heuristics and hope
Robotics Developer   11/21/2010 2:35:27 AM
NO RATINGS
I could only dream to be on ANY project that was on time, on budget, and up to full performance. I (when I was young and naive) would give realistic schedule estimates to my boss; he would double or triple it and give it the the VP; the VP would talk with marketing and come back with a schedule that was 75 to 80% of my original one. There is always pressure to shorten the proposed schedules either by the boss or higher ups. What usually happens is the engineering staff gets hammered and is blamed for the slipping schedules. NO-ONE remembers the original schedule, its just a fact. Until companies realize that the designs will take just so much time / resources and make the appropriate trade-offs UPFRONT they are bound to lose market share or fail. I might suggest a sliding schedule analysis that provides the cost verses time to market trade-offs balancing shorter/more costly development with earlier/more profitable releases. Anyone know a company doing this?

krisi
User Rank
CEO
re: Facts and data versus heuristics and hope
krisi   11/19/2010 11:45:31 PM
NO RATINGS
Ronald, you write "Projects miss schedule when management underestimates or fails to acknowledge the time and resources the R&D organization needs to develop complex ICs" but my experience on many IC projects has been that all projects miss schedule because they are designed to do so! As one executive explained to me, "if I make a realistic schedule we will be late to market by one year, but if I shorten it unrealistically we might be only 6 months late"...has anyone worked on an IC design project that met teh deadline and was on budget??? Kris

Radio
NEXT UPCOMING BROADCAST
EE Times Senior Technical Editor Martin Rowe will interview EMC engineer Kenneth Wyatt.
Top Comments of the Week
Like Us on Facebook

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
EE Times on Twitter
EE Times Twitter Feed
Flash Poll