 Reforming Project Management |
||
|
|
||
|
Thursday, September 05, 2002
Add Headlights to Your Project
I've spent the last three days working with a program management team who have ~80 projects to complete in the next three years. The scope is big; budgets are tight; they're planning a shotgun start on 4-10 projects; and staffing is not in place. Oh, I forgot to mention these projects will be performed throughout the US. Historically, this group and most other companies would hole-up to do detailed planning that would be base-lined at a task level. This would allow a central control of the project. The good news is this group of people have concluded the usual approach simply would not work. In no time the central plan would be out of sync with the reality of the field. This program requires a massively distributed project planning and execution system. The key challenge is to know that project managers everywhere are planning reliably. They have settled on a lean approach to project delivery using the Last Planner™ System. [The LPS is a trademark of the Lean Construction Institute.]
So, what's so special about this approach? To start, the focus of the system and practices is to make work ready -- that as tasks are coming up to be performed all constraints to performing the task are resolved prior to the start of the task, e.g., availability of competent staff, materials, tools, specifications, authorizations, etc. The intent is to get the task in a condition that it is ready to start and finish...in lean parlance once the performer starts work work continues until it is finished avoiding the usual wastes of demobilization and remobilization, waiting, and hand-offs. Earlier I wrote about should-can-will planning. This step is the "can" activity. The next difference is organizing recurring planning conversations with those people -- last planners -- who will assign the work to performers. In this conversation the last planners make promises to complete tasks on specific days. Planning-as-promising results in dramatically increased reliablility of task completion. I have seen overall project reliability jump from under 50% tasks completed as planned to over 75% tasks completed as planned just by conducting these weekly planning conversations with the last plannners. Finally, task completions are tracked just like Microsoft now recommends (see entry on 8/26/02) rather than tracking hours applied to tasks. Simple 'yes' or 'no' records are kept. Whenever a 'no' is recorded a reason for that variance is also identified. Accumulated reasons for variance can be studied and addressed so that over the life of the project some of the more prevalent causes of the variances are eliminated. Now, back to the 80-project environment...instead of operating under the illusion that a complete master schedule gives the program manager control over the projects, the program manager in this lean environment is monitoring the quality or effectiveness of the planning system through the results that are recorded on a week-to-week basis. This offers the opportunity for the program manager to offer assistance to those project managers whose planning reliability is low or not improving. One colleague describes the process as adding headlights to the project where the usual project only has taillights: after-the-fact end-of-month reporting. Visit the Archives for more postings |
 Reference Papers
|
|
| ||
