Drag and Drag Cost: How to Use Them
Stephen Devaux
President, Analytic Project Management; Author, Instructor, & Consultant
Got into a discussion thread about whether drag computation adds anything.
I'd suggest that:
So the exercise above consists of just 13 activities. I urge you to try timing yourself. How long does it take you to answer the answer the questions correctly?
Then try thinking what it would be like to get that same info for a medium-sized project of, say, 1,300 activities. Without software that computes drag.
And then doing it again when something changes.
As usual, I'll post the answers in the Comments section below next weekend.
President, Analytic Project Management; Author, Instructor, & Consultant
2 年So no one seems to have posted answers to the above exercise on critical path drag and drag cost. I don't know if some got the right answers but didn't want to post them; or some tried to do the computations but couldn't (which would confirm my feelings that Primavera and other s/w packages need to be computing these data!). Anyway, here are my answers: 1. Drags: A=7; B=2; B-E lag = 2; E=4; H=3; J=12. 2. Drag cost: A=$450K; B=$200K; B-E lag=$200K; E=$300K; J=$700K. 3. (a) Project is shortened by E's drag of 4d; (b) which is worth B's drag cost of $300K; (c) now drag of H = 5d. 4. It took me 1.5 minutes to compute the drags; 4 minutes to compute the drag costs; 3 minutes to figure out all the answers to 3. It would have been faster, but as it was my exercise, I had to check and make sure my answers were correct. If anyone has any questions/issues, I'll be happy to respond.
Developing IOT solutions for harsh environments | Founder and Director of MDDSystems Ltd
2 年I know it's not what you asked Stephen but the diagram does raise some questions. I very rarely use lags in my schedules as it masks what is happening. For example what are the lags for Tasks B and K? I'd ask the person who knows most about these two tasks; "what is preventing you from starting them on day 1?". They'll probably tell me that they have to wait for something to be completed in Task A and so I'd ask them how would they know when to start. There's probably a subtask within task A that is driving the start of B and there's probably a "milestone" that initiates when task B can begin. Hence I would split A into several tasks. The lag for B starting would then be a task and I can decide if I want to try to reduce the time of this task. I could put more people on this subtask or do shifts or do something to speed it up. Even if it is "drying time" then, in theory I can reduce this if it makes sense overall. A fixed lag on a schedule doesn't tell me what I can do to reduce it, in fact it looks like I can do nothing. The subtask within A is on the critical path and not the whole of task A. The same applies to the dependency between C and F. There's possibly a FS relationship between the last 10 days of task F and task C?
Developing IOT solutions for harsh environments | Founder and Director of MDDSystems Ltd
2 年I can do small projects like this using Excel....takes about 10 to 15 minutes to model them but I wouldn't like to do a large project that way. My Python software is almost ready and I need to test it with lots of different dependencies etc. It seems to fail on task B in your project above so I'll check why. It calculates all the ES,EF,LS and LFs correctly. Oh and I think the LF for Task A should be 12, not 14. My Excel model says A, B, E, H and J are critical and their drags are 0,2,4,3, and 12.
Planning and Scheduling Expert, Efficiently Planned and timely delivered several billion dollars of projects
2 年The question is not framed correctly. If duration is more the cost will be more. The concept of critical path is not used in this calculation. If we see critical and near critical path the answer is readily available there even without any calculation. There is no need to calculate drag and create confusion. Critical and near critical path is a direct and easier way.