Guild of Project Controls: Compendium | Roles | Assessment | Certifications | Membership

Tips on using this forum..

(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.

MSP S-Curves - Early&Late

30 replies [Last post]
Conrad Cannella
User offline. Last seen 12 years 30 weeks ago. Offline
Joined: 28 Nov 2008
Posts: 17
I’ve testing the macro s-curve downloading from Freeware code produced by Rod Gill, Microsoft MVP for Project but this tool don’t create late curve but only early curves. Somebody know how same tools/macro excel that create early and late curves?
Thanks!!!

Replies

Adolfo Caccialanza
User offline. Last seen 4 years 42 weeks ago. Offline
Joined: 3 Oct 2009
Posts: 14
Groups: None

The best solution for Progress Curves is at this site:

 

http://www.cla-it.eu/TMSDetails-general.html

 

Best regards

Jean-Yves Moine
User offline. Last seen 3 years 3 days ago. Offline
Joined: 17 Jan 2009
Posts: 87
Groups: None

With the method of "S curves", you build baseline ealry and late S curves with the schedule, and then you measure the physical progress without taking in account the dates (just weight and % physical complete). It's another method than yours. Yes, I do not construct the remaining S curves, I have the updated schedule which gives me the final estimated dated, with critical path method, the longest path give the final date. The S curves are built to measure trends, delay, float and physical progress. I let the schedule bringthe final estimated date and float. I don' t need your forecast curves, because I have the schedule.

Rafael Davila
User offline. Last seen 10 hours 16 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

I am not using EVM for my S curves, I am using the methodology the construction industry uses on regular cummulative S curves. We use contract cost (BOQ) as a mesure of progres, at times we use estimated manhours as a measure of progress, at time we use BOQ volume of work when the unit of volume of work is common such as CY of concrete on a particular concrete Phase.

EVM yields its own S Curves.

But as many already know there are certain metrics on EVM that are wrong. I do not like EVM as it is currently applied, in any case Earned Schedule.

Earned Schedule also yields its own S Curves.

http://www.google.com.pr/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rj...

http://www.google.com.pr/url?sa=t&rct=j&q=&esrc=s&source=web&cd=5&ved=0C...

Jean-Yves Moine
User offline. Last seen 3 years 3 days ago. Offline
Joined: 17 Jan 2009
Posts: 87
Groups: None

Rafael,

You put the ETC connected to the EV...

Do you understand that theorically it is false?!

BR, JYM.

Jean-Yves Moine
User offline. Last seen 3 years 3 days ago. Offline
Joined: 17 Jan 2009
Posts: 87
Groups: None

Rafael,

In the EVM there is the ETC (Estimate To Complete), this is your curves (projectiosn at terminaison, early and late). But the ETC is the extension of AC (Actual Cost) in the EVM, not the extension EV (Earned Value). "S curves v2" calculates the EV (in volum of work) with weights and % physical complete information. For you Actual = Actual Costs, for me Actual = Earned Value; it is not the same, that is why I say that your method is so bad.

Best regards, Jean-Yves Moine

Rafael Davila
User offline. Last seen 10 hours 16 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

We use software that can model remaining activity duration based on remaining volume of work and resource productivity on activities that behave in such way. Other activities are duration type such as concrete curing, their progress will be determined with passage of time.

On Volume of Work activities we compute progress based on actual volume of work, it is synonym of physical progress. The remaining work have remaining physical work still to be performed, projected with an early activity performance and projected with a late activity performance. It is very simple and even ancient software such as P3 provides for it, no need to use less functional external software.

I am not saying that your method of computing progress is bad I am saying that the curves are missing projected progress as per whatever method you use. It is not only about baseline projected progress for the whole duration it is also about current schedule actual and projected progress after DD.  You got to expand your method to display what is the projected early progress and projected late progress of current job up to the projected end date.

Jean-Yves Moine
User offline. Last seen 3 years 3 days ago. Offline
Joined: 17 Jan 2009
Posts: 87
Groups: None

Rapheal, As you say "The missing curves tells you what is the predicted cumulative resource/cost usage against time"...

So, it is not physical progress S curves, it is expenses forecast in terms of resources/cost...

The tool I am selling "S curves v2" calculates the physical progress, it is not the same thing.

sorry but your method is very bad !

Jean-Yves Moine
User offline. Last seen 3 years 3 days ago. Offline
Joined: 17 Jan 2009
Posts: 87
Groups: None

The spreading of a physical progress is NOT linear (miletones, technique 50-100, etc.)...

Your curves are false if you use % physical progress, ...

It's a poor indicator, and false!

BR, JYM.

Rafael Davila
User offline. Last seen 10 hours 16 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

No matter what method you use to calculate progress, after DD your project if not finished will still have some pending activities, and these have early and late dates that will intersect on projected completion date. The curves on your reference is missing the projected curves form the remaining work, the most important of the set. They only show past history, the baselines that where initially determined and actual that at DD is past history.

The missing curves tells you what is the predicted cumulative resource/cost usage against time, they display projected final resource/cost that can be different from baseline as well as the projected end of job that can be different from baseline.

The curves shown in your reference do not show projected slippage and therefore are of poor value.

Jean-Yves Moine
User offline. Last seen 3 years 3 days ago. Offline
Joined: 17 Jan 2009
Posts: 87
Groups: None

No Rafael, I am not agree because I use physical % progress on the schedule. 

If you use % duration complete you can do projections but it is a bad choice, wrong, not a good method.

How do you want to do forecast if you use % physical progress on the schedule's task?

You can see the expected finish date with the schedule and the critical path, with float.

your curves are nice but falses.

BR, JYM.

Rafael Davila
User offline. Last seen 10 hours 16 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

Jean,

The link only show Early and Late for Baseline plus Actual for current plan but misses the most relevant the Projected Early and Projected Late Curves, without these the set is useless. It is a poor choice for a sales sample.

S Curves are very simple, the following figure displays the full set, with the projected curves you now that is the projected cost and duration versus the baseline, without the projected you know no new prejections except past history.

S Curves photo SCurves.png

Best Regards,

Rafael

Jean-Yves Moine
User offline. Last seen 3 years 3 days ago. Offline
Joined: 17 Jan 2009
Posts: 87
Groups: None

you can buy "S curves V2" VBA EXCEL application: http://www.cubixpm-scurves.com/en/

it allows to :

  • Create 26 S curves (baseline early and late, physical progress) in automatic connection with a MS PROJECT or PRIMAVERA P6 schedule.
  • Spread the weights in a linear or Gaussian way over the tasks of schedule.
  • Create S curves with 3 criteria, on the basis of 5 task codes.
  • Manage a schedule up to 10,000 tasks.

BR, Jean-Yves Moine.

Rafael Davila
User offline. Last seen 10 hours 16 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

I agree MS Project goes better with the English language. Perhaps instead of ALAP shall be used ALAP-TF and ALAP-FF, different to Zero Total Float and Zero Free Float as to avoid confusion between similar names for different functionalities.

If you apply either ALAP definition (ALAP-TF or ALAP-F) to all activities you will get the same result as eventually this will cause all free float and all total float to be consumed.

I also understand some software are missing the Zero Total Float constraint, a constraint intended not to be used for scheduling but to be used temporarily for isolating float paths, a very useful functionality as long as you eliminate out-of-sequence events. Seems like Zero Free Float is similar to a Finish no Latter Than constraint but instead of using a fixed date it uses a calculated date causing the early dates equal the late dates of the activity.

Zero Free Float constraint is useful to model just in time deliveries.

I am also missing the Expected Finish Constraint, a constraint that causes the program to calculate the Remaining Duration in such a way that the “Expected Finish” date is met.(An ‘Expected Start’constraint is similar to a ‘Start On’ constraint.)

All software should provide all, it is up to you to make good use of these.

Zero TT and Zero FF

Trevor Rabey
User offline. Last seen 1 year 18 weeks ago. Offline
Joined: 29 Nov 2005
Posts: 530
Groups: None

The ALAP date constraint in MSP is not an error or a bug. It does exactly what it says it does, and yes it does suck all of the float out of the path.

A constraint that lets a task go to the end of its own float only would be the zero free float constraint such as is in P3/P6, but MSO has no such equivalent (fortunately, never really needed anyway).

Rafael Davila
User offline. Last seen 10 hours 16 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

Sory I no longer have the file as I am not a regular MS Project user I erased them long ago. Erasing from rapidshare was perhaps an error I did not noticed. I try to keep forever all my postings for the free use of PP members.

Well, the following tip I got long ago might be of some help for other graph needs.

MS Project tip

Rohit K
User offline. Last seen 3 years 38 weeks ago. Offline
Joined: 2 May 2008
Posts: 73

Rafael,

These downloads are not available on the rapidshare. Can you please reload them.

Rohit

Rafael Davila
User offline. Last seen 10 hours 16 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

DELETED - POSTED FILE LOST, NO LONGER AVAILABLE

Conrad Cannella
User offline. Last seen 12 years 30 weeks ago. Offline
Joined: 28 Nov 2008
Posts: 17
Thanks!
Rafael Davila
User offline. Last seen 10 hours 16 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229
Conrad

I take the opportunity to express my overdue you’re welcome.

If you do not want to set into global the macros, field formulas and the views you can still transfer these from the sample job to any of your existing jobs by opening both jobs and using Organize to transfer between the two opened jobs. This procedure can also be used as to repair any of these views or macros in case you have to, this without damaging your file or losing data.

My macros always return to the original constraints as to avoid erasing original constraints from your file if after running ALAP you run macro 01; this was on purpose, just in case you were wondering.

Best regards,
Rafael
Rafael Davila
User offline. Last seen 10 hours 16 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229
Thanks Bo

I will make a manual run on ALAP as my macro after running the ALAP and creating the ALAP Baseline then runs the original constraints in order to return the schedule to the original condition. Originally I had my macro to run ALAP without running back original constraints, it behaved as I was expecting but maybe the testing schedule was too simple to show what you say. It would be of help for our knowledge if you let us know about a few conditions where the unexpected behavior occurs, is good to know the traps, all software do have a few.

Unfortunately P3 is showing its age and P6 is a monster. I tried Powerproject and Spider Project and found very interesting functionalities like; in Powerproject multiple bars in a single line useful for scheduling multistory buildings and the ability to categorize links, useful to identify "soft" links, in Spider Project Linear Chain Reports, useful for road construction, these among many other improvements.

I cannot live with the fact MS Project do not allow you to control unique task ID and the limit of a single predecessor relationship between two tasks, this and many other missing functionalities that for some reason Microsoft people never understood and most probably never will.

MS Project is not the software of my choice but at times I have to deal with it, so for me it is a must to have some knowledge on its use.

The idea of providing access to the templates is not only to provide some assistance to others but to get it from others, for my own knowledge, from many of you who will warn me of any error or omissions. Your comments are exactly what I was expecting; therefore I will thank you again.

Any suggestions will be welcomed; even better, any improvement on the template will also be welcomed.

Best regards,
Rafael
Bo Johnsen
User offline. Last seen 8 years 4 weeks ago. Offline
Joined: 28 Feb 2006
Posts: 119
Hi Rafael,

I’m aware of this, but thanks for the short and very precise description of S-curves, their use and slight hinting to EVA.

I was only stating that when using MS Project one does not get the correct result as the ALAP constraint here doesn’t do what it should do.

When doing S-curves (using ALAP) and/or cost/resource load programmes I tend to use other software planning systems such as Powerproject or Pertmaster (maybe in combination with Excel where appropriate) as I find them better suited for this purpose.

Regards,

Bo
Rafael Davila
User offline. Last seen 10 hours 16 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229
Bo Johnsen

S-Curves are plots of forecasted versus actual accumulated cost over time. This is used to compare actual earned cost versus forecasted.

If you use early schedule forecast the curve will be generated from an early schedule, if you use late schedule the curve will be generated from a late schedule. The actual performance should be somewhere in between both curves as you might use some available float without delaying your job. Using only the early s-curve might mislead someone as your job being behind schedule when it does not necessarily is.

If in addition to ES, EF and Actual s-curves you also show the planned/forecasted s-curve you will also get the projected finish date showing how ahead or behind schedule your job is.

It will not show you the details nor highlight the status for intermediate milestone dates. This is a complement of your standard CPM reports and not a substitute; it gives you in a single Birdseye view your job status. It is a standard well understood by all in the Construction Industry.

Best regards,
Rafael
Conrad Cannella
User offline. Last seen 12 years 30 weeks ago. Offline
Joined: 28 Nov 2008
Posts: 17
many thanks!!!!!

Thank you !!!
Rafael Davila
User offline. Last seen 10 hours 16 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

Enclosed please find revised files as to transfer the macros that were set as Global. Please excuse any inconvenience as I am not proficient with MS Project, I just recorded some macros using the Macro Record function similar to Excel.

DELETED - POSTED FILE LOST, NO LONGER AVAILABLE

The prior link is deleted from rapidshare as to avoid confusion.

Best regards,
Rafael

Conrad Cannella
User offline. Last seen 12 years 30 weeks ago. Offline
Joined: 28 Nov 2008
Posts: 17
I know the MSP’bug but if you pay particulary attention when the activity assigned constraint ALAP you can reduced this problem.
But if you doing it you, throught the tool that I search in this post, you can quickly define early&Late....

Help me with this tools excel....!!!

thanks
Bo Johnsen
User offline. Last seen 8 years 4 weeks ago. Offline
Joined: 28 Feb 2006
Posts: 119
Hi,

Maybe a silly remark, but why bother calculating anything with ALAP when using MS Project, as the ALAP function in MS Project absorbs the total float of the activity assigned with the constraint ALAP, thereby delaying all its successors, even though these successors are still assigned the ASAP constraint.

The ALAP function is only meant to absorb the free float of the activity assigned ALAP and don’t cause any other activities to move in time. Using ALAP in MS Project will only cause problem.

I’m using MS 2000+2003, so I don’t know if this error has been corrected in MS 2007. If so, please discard this reply.

Regards,

Bo
Conrad Cannella
User offline. Last seen 12 years 30 weeks ago. Offline
Joined: 28 Nov 2008
Posts: 17
I’ve download this file and when I run Sample Early and Late S-Curve Job.mpp the project don’t open any macro and when I push alt+F8 (for open macro-windows) I don’t see the macro’ name that you wrote (SCURVE01_Set_ASAP_ALAP,SCURVE02_Set_ASAP_to_BL1,SCURVE03_Set_ALAP_to_BL2)....

How????

Thanks for your interesting...
Rafael Davila
User offline. Last seen 10 hours 16 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

Enclosed please find two files for the preparation of S-Curves using MS Project.

DELETED - POSTED FILE LOST, NO LONGER AVAILABLE


Best Regards
Rafael

Conrad Cannella
User offline. Last seen 12 years 30 weeks ago. Offline
Joined: 28 Nov 2008
Posts: 17
I too but I’ld reduced this procedure and create this operation more quickly without change ASAP but all automatic between macro excel...
Paul Harris
User offline. Last seen 1 year 6 days ago. Offline
Joined: 11 Apr 2001
Posts: 618
I normally change the ASAP to ALAP and run the S-Curve again.

Paul E Harris
Eastwood Harris Pty Ltd, Melbourne, Australia
Planning and Scheduling Training Manual & Book Publishers, Consulting and Training
www.eh.com.au
www.primavera.com.au