Browse Prior Art Database

Method and System for Scheduling Uncertainty

IP.com Disclosure Number: IPCOM000014178D
Original Publication Date: 2003-Jun-19
Included in the Prior Art Database: 2003-Jun-19
Document File: 2 page(s) / 41K

Publishing Venue

IBM

Abstract

A pervasive problem in the business world is the scheduling of projects with major uncertainties, such as research. Often funding for projects is predicated upon the timeframe in which the dependencies must meet with success based upon a ROI calculation. Current state of the art relies upon risk dependency identification and adding a 'pad' factor of time to adjust the schedule. What amounts to a 'best guess' scenario in the project management planning. The two major methods employed in scheduling and implemented in scheduling tools are; Padding, and Critical Chain Scheduling (This is a more sophisticated variant of simple estimation and padding.)

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 1 of 2

Method and System for Scheduling Uncertainty

The approach to solving this problem is two fold: First by creating an object that models each schedule item and aggregating it with a set of risk dependency objects (one for each identified risk) which all feed a percentage factor of confidence in meeting the commitment. Secondly by creating an interactive system which collects feedback (lack of response is also a critical measurement) and continually updates the overall evaluation of the project risks/success.

In addition to each task being given a risk level, each performer can be given a performance value and milestones can be determined by some value of probability.. and the outcome (end date and possibility) of success can be run time calculated as we progress we can provide an answer (within some limit of acceptability).

Project success can also be gaged by a number of factors, akin to 'how successful do we need to be?' The level of investment can be adjusted based upon the probability or projected degree of success. (A method to get around the 'throw money or people at a project' reaction to slipping deadlines as exemplified in the "Mythical Man Month".)

Below is an example of typical research short term scheduling and the types of problems this approach will address:

"We schedule things correctly in the short term (e.g. 1 month), since no matter how uncertain research is, people tend to have a pretty good idea about what they will do when they show up at work tomorrow. This tends to be okay for a period of 4 weeks or so. But then, all scheduling beyond that has proved futile. Things simply change, people discover things, they branch to what they consider to be greener pastures, and unknown problems may crop up in the middle of the work and derail any scheduling that you may have. So, what we have done here is to try to schedule in the small, right now on a quarterly basis, and effectively put another schedule for the next 3 months and so on. I admit that 3 months is pretty long for the purpose of this discussion, but we can't afford to go through the exercise every month. Throughout, we also maintain the main objectives of the work (which tend to shift and change as well) so that we don't lose track of the big picture."

This method has advantages and disadvantages. One advantage is that people know pretty well wh...