TECHNOLOGY MANAGEMENT COUNCIL

Presentation slides from the meeting.

Presentation slides from the meeting.

Presentation slides from the meeting. More info on www.mironov.com

Presentation slides from the meeting

User passwords are stored with strongest encryption.

Presentation slides from the meeting

Presentation slides from the meeting

One of the issues that several people raised was the applicability of value-based scheduling to research groups. Ilya and I chatted about this briefly after the presentation, and we had a quick e-mail exchange as well. Here are the points that he raised:

I really liked your presentation, as well as I found the value-based
idea for project management as fresh and very promising. I do have a
concern of effectiveness of this idea in startup environment with a
significant degree of innovation and R&D present. The projects tend to
be less defined, there are more open-ended tasks, and potential market
analysis is very murky. I am curious to hear your thoughts of how your
system may address these special cases.

For these cases, there are a couple of techniques that should be used in addition to value-based scheduling:

• Workflows: Any research activity will have false starts and dead ends. The key to being effective is having appropriate decision points for checking progress and deciding whether to iterate, try something new, or stop. Using these workflows, we can automaticallly adjust or even redefine the project structure when we reach these decision points. Value-based scheduling can work with this automatically (that's is how it's implemented in Lakeway Frontier).
  Any research activity will have false starts and dead ends. The key to being effective is having appropriate decision points for checking progress and deciding whether to iterate, try something new, or stop. Using these workflows, we can automaticallly adjust or even redefine the project structure as we reach these decision points. Value-based scheduling plugs can work with this automatically (this is how it's implemented in Lakeway Frontier).
• Buffers: Someone in the audience asked how the "critical chain method" could work with value-based scheduling. I think this is a case is a good illustration. In the critical chain method we recognize that resource bottlenecks are what often drive schedules, but because they change so quickly, it's hard to know where they are. The "workaround" for this is to estimate tasks at 50% confidence and 90% confidence and then sum the difference between these estimates (for all tasks) to come up with buffers for the project (effort, budget, duration). We set the target date for the project to be the 90% estimates, but we manage to the 50% numbers. Value-based scheduling comes in by computing marginal value at both the 50% and 90% dates and combining them to yield an "expected marginal value".
• Expected product value: Research is tricky because often the outcome is binary: it's worth $1B or nothing. However, we start these projects because we have some understanding of the risk. We also start these projects because we have some understanding of the market. If we have enough information to approve these projects, we have enough information to come up with value curves. They may change more radically as the projects go on, but that accurately reflects the situation and the information we learn as we go forward.

Wow, it's a big step forward for our site.  Thanks for making this happen.  :-)

Greetings,

I've created a page for venue booking information:

http://www.ieee-scv-tmc.org/newsite/board/venue-booking-instructions

Cheers,

Chris