Dear Sir/ Ma'm/ Friend,
(1) How does the limited resource (labor, material, equipment, money) allocations are carried-out among the multiple projects?.
(2) Do they (planning engineers/ resource manager/ PMO) follow any set of heuristics to optimize the project time, and cost in multi-project environment?.
(3) Does the available project management software (P6, MSP, any other) covers this multi-objective, multiple resource management in multiple projects?.
Finally,
(4) Does any of the project management software's has the ability to optimize the objectives (optimization algorithm, multi-criterion decision making algorithm)?.
Expecting your valuable comments.
Thanks.
Marimuthu K
Dear Marimuthu,: the tipical model for many Projects and limited resources is a LP model or other OR model. You can solve the second Problem thhrough Simulation or PERT. Complex Systems, multiproject, multiresources, multiobjective can only be solved by Simulation. The existing Software calculates an objective/multiobjective but does not optimize at the same time. Regards, Ricardo
Yes. you rightly pointed out that "simulation" could be the better solution to predict the complex situations. And by the way, from the literature it is possible to obtain the Pareto front for the multiple objectives through multi-objective optimization algorithms (GA, PSO, SA, ACO). Further it can be evaluated by MCDM methods. And my question is that does any project management software has this facility of doing this so?..
Dear Marimuthu, I don't think there is any algorithm for complex Problems. For simple Problems you should just use the formula developed for a particular MCDM (for example compromise programming) and obtain the Pareto optimal Solutions and sensitivity Analysis for several weighting factors. Regards, Ricardo
Thank you for initiating the discussion and valuable information.
There are some heuristic and meta-heuristic algorithm for solving these kind of problem. And as Ricardo said, there are few OR-based model to formulate these problems but this is an NP-Hard problem meaning that they cannot be solved for medium size and large size problems. Simulation is the best tools to test heuristic algorithm and also for scenario-based optimization.
A Divide and Conquer paradigm such as cooperative Coevolution (CC) has recently been viable solutions for multiproject, multiresources, multiobjective problems 😎
Dear Marimuthu,
You may have a look to IBM Ilog Cplex CP Optimizer. CP is for Constraint Programming and O for Optimization. The modeling language contains a temporal language for (almost) all practical time dependent combinatorial problems. That is a stand-alone black box solver so you can include as a scheduling generator in your own application.
- Badges
- Science topic
- Similar topics
- Computer Communications (Networks)
- Simulators