10-4

Hi Rafael

Why restrict your campaign to "get mike" to just yourself and Vladimir.

Let us open it to the whole of PP.

A new thread is started in the Training section.

Please join in.

Best regards

Mike Testro.

Vladimir

You don’t have to get into strictly linear project to have linear chains. It is very common; I counted 26 such occurrences in a 450 activities schedule, while justified use of negative lag is really uncommon though.

Because I did not overlapped enough activities I got in my updates a lot of out-of-sequence occurrences in my FS0 relationships. Progress overrides was the option that best reflected the change in plans and its effect in the following activities, another “forbidden” schedule set-up, they don’t get it, none is a sure fix for out-of-sequence occurrences.

AACE even recognizes the possibility of negative lag and accounts for it in their procedures. They use lags to their advantage, most probably this makes the procedure easier to program into computer software.

Best regards,
Rafael

Mike,

How about a work package first start always being a Monday? Is a piece of cake, understanding what Andrew is looking at, is serious business.


Remember if too small for your view you can zoom your PC screen.

Vladimir

No matter how many banana peels I throw on the floor Mikes gets away. Do you have something new to make him fall?

Best regards,
Rafael

Vladimir,

Think of the one that will get Mike into trouble, at some point the overlapping even in a simple network will create an incomprehensible schedule with activity splitting.

Mike,

All the lead lags would be set to Calendar Days.

Ironically I cannot do that unless I create an additional dummy activity in between as my software does not let you choose the calendar for your lag, it assigns to the worst of choices, to the predecessor activity’s calendar, and you know it.

I never go into the granularity of concrete curing and use only one calendar for work days with the exception of a single bar, a hammock that spans all job duration, this I report using Calendar Days.

That was just an example, we all know you bath in baby oil and avoid showing us your solution to our scheduling needs, you will only show your own. If you don’t like the concrete curing sample let it be Wall Paper, Activity 1 = Glue on Wall, Activity 2 Place Paper, Activity 3 Roll Press Wallpaper, Activity 4 Clean Excess Glue. It can be any linear process, say cast in place terrazzo floors, say vinyl tile installation; too much granularity then, let it be Pre-stressed Concrete Pipe Installation on the middle of an existing highway.

Best regards,
Rafael

Hi Rafael

You have hit on the one set of circumstances where I use FS lead lag links - namely when curing or drying out is going on.

The links would be as follows.

1. Pour Concrete > FS link to
2. Place Burlap > FS Link to 4. Remove Burlap
3. Pour Concrete > FS lead lag of 20 days to
4. Remove Burlap

All the lead lags would be set to Calendar Days.

I use lead lags because if there was a physical task representing the curing time then I would have to estimate progress by way of the number of days elapsed whereas the Lead Lag just gets rolled up.

I had this disussion with Andrew recently which you may have followed.

I have yet to discover a way of retaining the upstream critical path if the calendar lead lag ends in the middle of a holiday.

Best regards

Mike Testro

Hi Rafael

I have had a quick look at Fig 7 which represents an hypthecal situation.

Most contractors programmes are full of devices that obstruct dynamic change (see Ban these planning abominations)so the first thing I do is to clear them all out so as to create a realistic critical path.

I do this before I have investigated any of the delay events so that I cannot be charged with rigging the baseline programme.

The next stage is to develop a level of detail down to level 4 so that events can be impacted with greater accuracy.

Then I can start on a realistic analysis.

Best regards

Mike Testro

Hi Vladimir

I don’t know if this works in Spider but in Powerproject a FF link takes on the calendar of the task at the end of the link.

Thus if you set up a FF link from a 5 day week task to a 6 day week task and set a negative lag the same as the 6 day task duration then the critical path is maintained and it looks as though you have a FS link.

I personally would not bother and I have not yet tested what happens to the logic when the 5 day task is 100% complete.

Best regards

Mike Testro

Hi Vladimir.

Precisely.

So if an activity is set to a 24 7 calendar and it is linked to an 8hr day 5 day week activity and the first activity ends on day 5 of a 10 day holiday then all upstream tasks will have 5 days extra float.

As our Russian meerkat would say - "Seemples".

Best regards

Mike Testro

Hi Andrew

I have come a bit late into this debate but I would respond to your thread #1

1. Can anyone think of why you should replace float, (a variable time period), with a lag, (fixed time period)?

Yes to allow for a curing or drying out period with a lead lag link with a set number of calendar days - between say > Lay Screed > Lay limestone flooring - 75 calendar days.

This is the only circumstance that I would allow a lead lag into any programme.

The problem is that if the 75 day lag date ends in a holiday period the float is increased by the number of days before the holiday ends.

This could take the upstream activities out of the critical path.

I have not yet found a way in Powerproject to ovecome this problem and keep the dynamics of the logic.

Best regards

Mike Testro

Vladimir

For the record;

I do not believe in artificial links or “soft links” even in the small schedules I get involved I lose track of them, the problem is that some Contractors don’t use the functionality of resource constraining.

I believe any procedure that does not take into account the effect of resource leveling is not valid.

I do not believe in any methodology that uses a Baseline Schedule or “As Planned” as a reference. Logic changes, and scope of work changes, almost everything changes and you have to account for all changes. I would suspect all As Planned Schedules to be a lie in a claim enviroment.

I believe the forensic procedures should be based on methodologies that can simulate to the best extent what really happened as decisions were being made, in the prospective, irrespective of whether the submitted schedules were lies, half lies, full lies or true lies, no matter what still lies. I don’t like the idea of competing procedures claiming to be the best, this claim alone makes me doubt about all procedures. Until this happens, if it ever happens, we will have to do the best with what we have on hand. This is not Disney World.

The true question is, is if the procedure is valid, not if is perfect. I believe this is the essence of Andrews’s question, a very good question; I don’t have yet the understanding on the methodology to help him validate or reject his point. It is a 118 pages technical document written for the experienced forensic analyst.

Best regards,
Rafael

Rafael,
we incorporate in Spider Project any idea that looks reasonable. So I usually suggest our clients to ask questions if they could not find in Spider Project some functionality that is needed. I am sure that can explain how to.
But adding links between activities that create a sequence in the current schedule due to resource restrictions is obviously poor idea. Just imagine construction schedule consisting of many thousand activities. Some of them were delayed because of resource, supplies and/or financial restrictions. You shall find which activities were postponed for which reason, find activities that use the same resources and finish earlier, understand if the delay was caused by some of assigned resources shortage, or by material shortage, or by the lack of money to proceed. The last two options are most problematic and later I will explain why.
Then link all recent activities that use the same resources with the postponed activities with FS links and mark these links as artificial. These links shall be added only to those activities that will start next week because you just don’t know what order of activities will be in the revised schedule calculated a week later. A lot of work and very hard analysis later - many thousands artificial links that showed only that some activity was postponed for some reason.
Delays due to supply and financial restrictions are most difficult to simulate through links. If your schedule includes supply (or financing) activities/milestones the software knows when next portion of materials/money will be available. So activities are delayed until planned supply will come at the moment when material/money will be spent. It cannot be simulated as FS activity. It is FS link with negative lag from supply/financing activity. These links with negative lags shall be added to all delayed activities that start before and finish later then financing activity. For materials I shall select and link only those delayed activities that use the supplied material.
Imagine the number of these silly artificial links!
What you will do with all of them in delay analysis?
Best Regards,
Vladimir

Vladimir and Andrew

I don’t think the Protocol necessarily disagree with any of you, most probably agrees both of you are right. It recognizes the practical limits of the various methods and provides a listing on the disadvantages of each method at the end of the discussion of each one. Most probably the particular procedure you are analyzing has a disadvantage with regard to prospective analysis and the limitation is named in the protocol, it is too complex to memorize all limitations for each particular technique at first reading.

The document warns the reader that it has to be viewed as a whole and is not intended for the novice in forensics, this rules me out of any relevant contribution. In order to understand the Protocol I don’t believe it is enough with reading it as if a novel, you got to do what Andrew is doing, running yourself the examples on your own computer.

If I were to take all the effort into any procedure it would be in a procedure that can reflect changes in the prospective critical path indicated in the contemporaneous schedule updates. I would have a preference on the method by Ron Winter’s Schedule Re-Builder whenever it can be applied. The use of computer software greatly reduces the effort to apply the technique, especially on the hands of an expert.

"Schedule Re-Builder destatuses an As-Built Schedule and then rebuilds it update period by update period based on expected progress and past actuals. Re-Builder tells you what Project Management thought happened to the critical path at the time of construction. SA Profiler adds the ability for Schedule Rebuilder to consider the actual work progress for in-progress activities falling on the data date."

With regard to the issue of resource constraining, the Protocol opens the option to consider it; "Conversely, if resource constraint is at issue and the schedule logic does not reflect the constraint, insert resource-based logic to obtain a critical path that considers all significant constraints." Is not that you will be able to apply a computer resource leveling run, but it will model the effect.

Best regards,
Rafael

Yes, Andrew, I agree.

But I want to add that if delays were caused by resource restrictions then transfering them into lags and links is not correct also. In this case the cause of delays will not be easy to find.

I think that it is useful to recalculate the initial model with as built activity durations and compare the result with initial and as built schedules where as built means just actual dates and not artificial CPM schedule.

Best Regards,
Vladimir

Andrew,

I agree, by modeling the sample is not only the best way but frequently the only way, I will follow your suggestion. I know it won’t take long to setup the model, about the answer not sure, I am not into forensics and my knowledge is limited to the application of TIA in the prospective.

Best regards,
Rafael

Andrew

This is a methodology to converts an As-Built schedule into a unstatused CPM schedule. If you compare figure 6 AS-PLAN LOGIC WITH AS-PLANNED DURATIONS with figure 6-AS-BUILT LOGIC WITH AS-BUILT DURATIONS (RIGHT) there was an out-of-sequence occurrence.

The purpose of this is to allow the CPM schedule to simulate the actual activity durations and sequences solely by CPM computation using the logic ties and actual durations.

Because I only have experience with our definition of TIA and never used forensic methods I would not dare to comment on it, just to provide you with the best reference I can think of.

If you go to page 115 you will find that Ronald M. Winter, PSP is one of the authors, if he does not show maybe you can contact him through PP messaging system or through his web page.

http://www.ronwinterconsulting.com/

He developed software that can do that, a part of a Suite called Schedule Analyzer Forensic.

http://scheduleanalyzer.com/forensic_brochure.htm

As-Planned/As-Built Maker converts an As-Built schedule into a unstatused CPM schedule with modified durations and logic to reflect the As-Built condition. This As-Planned schedule will reflect both the early starts of activities as well as the actual durations. The logic change overcomes the objection of using actual durations for analysis. SA Profiler can work with schedules created by As-Planned/As-Built Maker to create an actual daily activity CPM network reflecting actual logic used.

By the way I love the following argument in the AACE document.

In most cases, simulating the actual performance of work using CPM logic requires the use of logic ties other than standard, simple, consecutive finish-to-start ties (FS0).

My Own: Those against use of logic ties other than standard, simple, consecutive finish-to-start ties (FS0) open up your minds. Sorry it was too much of a temptation, is for Mike and his gang.

Best regards,
Rafael