(1) Explain your problem, don't simply post "This isn't working". What were you doing when you faced the problem? What have you tried to resolve - did you look for a solution using "Search" ? Has it happened just once or several times?
(2) It's also good to get feedback when a solution is found, return to the original post to explain how it was resolved so that more people can also use the results.
Hello All,
I've been a project engineer for several years on the mechanical end of oil/gas and HVAC projects. Consequently I've been exposed to Project Management basics such as MTO, cost estimating, deliverables, milestones, Gantt charts, dependencies, CPM etc. The scheduler is already overloaded with projects (so I'm falling into this additional scheduling role so I hope my practical engineering experience will be helpful. Whiile my exposure to scheduling has been valuable, has also been informal. I am cognizant that there is a HUGE body of knowledge that I am unfamiliar with and I have the utmost respect for you scheduling professionals that make projects happen. I'm looking forward to learning :-) With that being said, here are some questions with regards to an oil/gas project:
1. Deliver a Area 6 week Schedule and Site 2 week Execution Schedule. My research shows that this may be related to the detail Level (1 to 5) you are dealing with but I'm not sure. Is a ""Area 6 week schedule and Site 2 week Execution schedule" a common time frame in project scheduling or project specific? I'm assuming the latter.
2. Execution schedules for CAP activities? My research indicates CAP refers to Control Account Plan and deals with activities to be performed as it pertains to the project budget and schedule.
3. Resource Leveling? It seems this would refer to optimizing resources (personnel, equipment) to perform the project activities/tasks.
I apologize for the long post. Thank you for any information or confirmation!
Regards,
Iz
Printer-friendly version
Send to friend
hello all !
I am new to Planning but know the basics as I did CAPM from PMI.
I am designated as planning engineer for 53 km road stretch which is to be double laned from existing single
in a hilly terrain Indian State.
Prepared L1 schedule got 6 main activities.
bush clearing, earthwork, sub-base, base, bituminous sub base & wearing course.
all activities need the previous one to start but not necessary complete for the whole 53km stretch.
but for A6 - A1 to A5 should be complete [ e.g. doing A6 only for 2km stretch which is practical at site]
how to i prepare a detailed schedule for such a case.
can these activities be counted as ladder activities ??? can i use Steps in P6 for any help.
or shall i make these 6 acitivites as a part of WBS & then break 53km into 5km stages, with each stage with above 6 activites
we do have many more activities but managing them is not a problem as they are not linear acitivities.
does any body has any idea of planning road works previously ???
plz help me as my job is in STRAIT.
plz e mail any template to study @ ss.dhatwalia@gmail.com or forward me some useful link.
thank you all.
Iz,
The following reference might give you an idea of what resource leveling is.
Part 3 is specific to Resource Constrained Scheduling.
The following reference is an introduction to how a scheduling software works in actual practice.
http://www.spiderproject.com/images/img/pdf/Spider%20Intro%202013.pdf
Best Regards,
Rafael
Wonderful information! Thanks for the great input Larry and Mike and to the rest of you who have given me a glimpse into some of the debates within this field.
Regards, Iz
Bodgan, regarding structural analysis algorithms you said - everyone knows that any of the methods applied produce at least acceptable results. This shall also be enough for scheduling where hundreds of tests can be done on the computer without fear of a bridge or building collapse.
http://www.softwaretestingstandard.org/part2.php
In any case software specifications should require the software to be certified as flawless with regard to resource leveling, float values and other basic requirements. The software shall be tested by different institutions, as to make the process unbiased, transparent and any user shall be able to challenge the test results by exposing any flaw they can find. The same rule shall be applied to resource leveled float and software incapable of consistently producing correct values of floats be banned.
Some resource leveling algorithms are documented and some yield better results, but the programming is dependent of specific software features, I know no other software capable of dealing with variable quantities resource leveling so Spider algorithms can only be tested using Spider. In addition the effort into developing and applying the algorithm is intellectual property.
That an algorithm is yielding a feasible solution is easy to prove, just expose the resulting resource assignments and look for overloads, if there are no overloads the schedule is feasible and the algorithm works. Try it with hundreds of scenarios and if it always yields feasible schedule then the issue becomes which consistently yields better results.
Because better results can be relative, Spider Project allows the user to select among a selection of algorithms and one that shall be considered is after selecting your plan using whatever algorithm fits well is the option that follows previous version activity sequence, this will keep some stability on your planning until you start becoming into trouble and need to take some action to get back into healthy probabilities of success.
I do not agree with those who claim the algorithm shall be exposed, usually software developers that favor weak software, or perhaps incapable of understanding how procedures of quality control can be applied to complex software. In any case that a software developer claims he used an acceptable and known algorithm does not means it was correctly implemented. Some software are notorious for not being able to yield basic float values, and the math is known by everyone.
Others who frequently have issues with resource leveling algorithms are old school analysts most notably those who publish cookbook delay analysis procedures. They pretend, to the detriment of claimant, to disclose every computation the software did when in civil cases it is enough to prove the plan is reasonable and feasible. But they have their own economic interests on the sale of their products and services.
Bodgan,
I agree with all your points!
About comparing use of computers versus manual methods, my response is a sarcastic response in agreement with your statements. I cannot understand why someone in the new millennium is not willing to make use of computer assistance to make their informed management decisions, same as a structural engineer use computers.
About never trusting your computer to screw up your programme I guess I provided a direct answer, the computer is a tool that must be managed. The computer is a machine that can be programmed within certain limits to make repetitive calculations based on pre-defined rules and that is all.
Good software shall be able to assist you to make resource planning, if not then you are using wrong software tool. Good software shall be able to suggest variable quantities and workloads and what impact it will have on activity durations, this is where you can differentiate the kids from the adults software. The software will still need your continuous input but better than doing it the ancient way.
Management is not all about pre-defined rules that never change, therefore we must learn to make use of computers to assist us but never pretend once the initial schedule is submitted nothing will change, we must continuously manage the model.
Among the "rules" imposed by some specifications I find the most damaging the one that restricts the contractor to make changes in plans, second might be the application of earned value that is so dependent on a fixed baseline that owner/client are reluctant to make changes in the baseline, an ever changing baseline with an ever changing critical path, this makes you loose vision of actual targets in favor of past performance, is insane.
Best Regards,
Rafael
Rafael,
I'm sorry but I do not undestand the point you are trying to make. With which point(s) of view do you agree/disagree, if any.
Regards,
Bogdan
Bodgan,
Forget about finite elements methods, forget about structural dynamics and earthquake, forget abut the possibility of resonance during an earthquake, forget about long hardy cross method use the short method of two cycle moment distribution and whatever goes on first floor goes on the one hundred above. Apply the same theories to space shuttle design, fly your stealth plane without computers, and navigate to outer space using a sextant.
The same goes with resource leveling, hire twice the maximum required labor to execute as many activities as logic alone allow, install a thousand cranes on your buildings, go back to the stone age if you want to because your sloppy software is not good enough to be of any help.
Idle resources is one of my major concerns so I use software capable of dealing with the issue of variable quantities and workloads, I am not to use software that delays an activity because 1 of 10 required resources is not available. Like it or not software resource leveling capabilities makes a huge difference.
I have noticed on the discussions about resource leveling [constraining] that many believe the limits must be real when the limits can be best used as a way to reduce idle resources and smooth peaks on resource planning in combination of variable quantities and workloads, these are management decisions, not computer decisions.
Considering resource constraints complicates delay analysis, therefore many analysts insist on avoid the issue and apply cookbook procedures as if baking a cake.
I will make my resource planning decisions with the assistance of good software, it makes no sense using software that is of not much help.
I 100% agree with your call - My standpoint is that I would NEVER trust a computer to screw up my programme - I need to be in control.
Especially with delay claims I believe your call is of most relevance, schedules must be managed and cookbook delay analysis recipes are flawed, this is not how jobs are managed. Unfortunately archaic analysts insist on applying the cookbook methods, most notably the AACE International that go as far as avoiding the issue on resource planning as if it does not matters.
Best Regards,
Rafael
Hi Bogdan
Computers have no brain.
If you like the idea a brainless idiot screwing up your project programme then please go ahead.
Best regards
Mike Testro
Iz Gee,
Here are some suggestions in preparing the schedule.
You need to establish the following, preferable to prepare them in Excel (beginner's approach);
-List of activities (Level 5 as you said)
-Duration of each activities
-Sequence of activities
-Logic relationship between activity
-ID Coding system
-WBS & ABS
-Start and Finish Milestones & Intermediate (constraints), if any
-Calendar (work hours, days, weeks, holidays, etc.)
After establishing the above, you can input and run them in Primavera.
Hope these helps...
Larry Rino
Mike, you said:
I am going to have to take Rafael's and Vladimir's side on this. Depending on the complexity and level of detail of the schedule, the ideea of manually leveling a programme using softlogic, in my view is the same as saying a structural engineer prefers to apply, for a 60 storey building, the Hardy Cross Method for determining Bending Moment Distribution (which is an iterative process) using a pencil and a couple of hundred pieces of paper instead of using a software designed to use this method.
Best Regards,
Bogdan
Hi Iz
I am not a planner but I think I can help a bit.
I assume you are working on P6 - anyway the principles are the same.
Area and Site are indeed most likely to be different programme levels - check your WBS.
The 6 week - 2 week requirements are look ahead programmes which is generated by the filter mechanism. The result should be all the level 4 tasks programmed to be done between the two dates.
CAP tasks has probaly been added to tasks as a code or cost value.
Once again you can use the filter to show them for the whole project or the time bands.
Resource levelling is a facility that requires the software to adjust your programme so that any selected resource does not exceed the stated deployment limit.
Thus if you only have three supervisors on site but they are required in 5 locations your computer will decide where they should be placed and adjust the end date accordingly.
Rafael, Vladimir and I have engaged in lively debate on this topic over the years.
My standpoint is that I would NEVER trust a computer to screw up my programme - I need to be in control.
So I generate a resource histogram on the Supervisor allocation and then adjust the logic to even out the peaks.
This is sometimes called "Soft" logic.
It is much easier in Asta PowerProject where you can allocate a name to the link and then switch it on or off.
I hope this helps
Best regards
Mike T.
Email: PPadmin@planningplanet.com
Replies