The Computer
Science Colloquium
Thursday, February 23, 4:15pm,
room 9204/9205
Murray Cantor
(IBM)
"Embracing Variance"
In order for most development projects to be funded, the management requires an initial
estimate of the time and effort required for completion. Estimates are not simple values, but random
variables best described as a statistical distribution; nevertheless most often only the nominal value is reported.
The variance of the estimate is a good measure of the risk of the development program and therefore is vital
information for development managers. Further, every entry into a Gant chart is an estimate with unreported variance.
Using minimal mathematics, I explore the implications of the fact that program parameters are estimates with variance,
and how the estimation variance may used to enhance the management of development programs throughout the lifecycle.
Further I introduce the notion that recognising and addressing the variances explicitly allows enables the organizations
to manage risk and create value.
Bio: Dr. Murray Cantor is an IBM Distinguished Engineer and a Rational Software CTO team.
His areas of expertise include software and system engineering processes, and system development management
and leadership. Cantor is the lead architect of RUP SE, the extension of the Rational Unified Process for
system and enterprise engineering. Cantor has been a system architect, team lead, project manager,
development product manager, architecture manager, program manager, and lead consultant. Throughout his career,
he has focused on the management of large-scale, cross-discipline development efforts.
He is the author of two books: Object-Oriented Project Management with UML,
published by John Wiley in 1998, and Software Leadership, published by Addison-Wesley in October, 2001
Cantor is currently collaborating with Scott Mathews from Boeing Phantom Works and Vasco Drecun from CPD Associates
on a program to extend Black-Schole financial calculus methods to development effort estimation and governance.
The Colloquium is supported by generous contributions from
the Bloomberg, Information Builders, Inc. and Netlogic,
Inc.
|