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.

Modeling of Negative Float with Spider

10 replies [Last post]
Rafael Davila
User offline. Last seen 4 hours 32 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

Although I do not find much value in the modeling of negative float it is frequently a requirement on our schedule specifications. Therefore I am looking for the best way to do so.
At the moment I am using a comparison schedule similar to current schedule but set up with all activities ALAP and a schedule run similar to the current schedule [regular/leveled] but backward computation. The Negative Float value equals the comparison difference field DIFFERENCE-Start. I am using bar colors defined with a signal so most critical are red, those with negative float but less critical in yellow and green for activities with float.
I am trying to polish my procedure so I am asking for suggestions.
•    Is my procedure correct?
•    Is there an easier way?
•    One of the issues I am having is on setting up the color signal based on the value of DIFFERENCE-Start field so that red is always only minimum value, yellow for values greater than minimum up to 0, and green for values greater than 0. For simplicity any suggestion that do not require use of formulas and scripts will be welcomed.
•    Another issue I am having is with the label of comparison fields, I am not being able to change them.
The following figure displays the logic I am using. I am showing Early and Late Bars for current schedule as well as Negative Float Bars from compared schedule. Also I should be able to display Separate S curves for all bars, something important to me as the substitution of current schedule late bars with the negative float bars yields non-matching sets of S-Curves that are misleading.

 photo negativefloat_zpsfb0134f6.jpg

Best Regards,
Rafael

Replies

Rafael Davila
User offline. Last seen 4 hours 32 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

Vladimir,

Thanks a lot it works as expected, without writing on top of real CPM float values, it is also transparent about it being Negative Start or Negative Finish Float as both are always displayed, contrary to other software you do not know if Start or Finish Float unless you verify the hidden settings.

 photo negativefloat_zpsa78c3b58.jpg

Best Regards,

Rafael

Rafael Davila
User offline. Last seen 4 hours 32 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

Vladimir,

About duration difference maybe it shall be converted to a common calendar duration and then difference shall be calculated. Which calendar I believe shall be at the option of the user.

Also a calendar durations calculator can be very handy, it will not be enough to say change duration from one calendar to another but the start and finish dates shall be considered as to account for calendar exceptions. You enter 1st calendar duration and start or finish date, you enter desired 2nd calendar and desired start or finish date on second calendar. Start and finish dates are related to duration so one will give you the other.

 photo spcalculator02_zps353f91c6.jpg

     ACTIVITIES DURATION OR WORK TIME IS DEFINED BY THE UNION OF ALL TEAM CALENDARS.

A bit more information such as work week data and exceptions within range can be shown in the print reports. Maybe you could be able to input&/output several calendars [CalendarA, CalendarB, ...] to find out the meeting calendar, good to consider impact of different resource calendars. In the same way we are concerned about meeting calendars for teams we are concerned about union calendars for group of teams. If the calendars are incompatible the tool will let you know on the spot.

I got from the web the following figure of Ron Winter Date Calculator which gives me a fraction of the information I am looking. Quite useful but the idea might be expanded for conversions from one calendar to another. Maybe the data as shown in Ron Winter Date Calculator can be shown under each of the two calendars on the proposed Calendar Conversion utility.

 photo date_calculator_zps29e3f89b.gif

BTW my request for Negative Float Calculation is to satisfy a common contractual requirement on our jobs for which I do not see much value, perhaps none and misleading as applied in most software. So after satisfying the need of the contract I will do nothing with the Negative Float values. Once times elapse management is not about what should have been but about what will be. We look forward and Negative Float obscure the forward view. Sorry for asking you for what I consider a misleading value but such is life.

On the other hand the functionality to calculate date differences using a selected calendar for comparison fields as well for formulas I find it of much value.

BR,

Rafael

Rafael.

don't forget that there is also duration difference. So comparing dates in two schedules you may also show duration difference. It will will show date fields difference in activity calendar.

Rafael Davila
User offline. Last seen 4 hours 32 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

Vladimir,

In addition to the modeling of negative float would it make sense for date fields difference be calculated on a set of options from Comparison Setup to be selected by the user? ie.

  1. [Main Project Calendar]/[Individual Activity Calendar]/[A selected Calendar]
  2. Calendars as defined from either [Current Schedule]/[Compared Schedule]
  3. Time unit [Days]/[Hours]

When the cursor passes over the DIFFERENCE column label the user selected options shall also be displayed.

BR

Rafael

Rafael Davila
User offline. Last seen 4 hours 32 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

Vladimir,

Thanks a lot. I believe real float concept is basic for the resource leveling algorithms be efficient and wonder if comparing resource leveled jobs with constraints would make the difference between software even bigger.

Best Regards,

Rafael

In the project proprties select the calendar for Average Work Days from the calendar field.

Thank you, I will look at requirements. It is not hard to create additional field for "properly" calculated negative float if necessary. As additional field it may be used without damaging real float concept.

Bars are correct and S-curve will be correct too. Only float values may be wrong if they are based on different calendar 

Rafael Davila
User offline. Last seen 4 hours 32 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

Vladimir,

After my last post I tried the following.

I created a new Calendar 2 equal to original and only Calendar 1 and assigned it to all activities. I changed workdays on original Calendar 1 to 3 days per week. But the negative float values did not changed, as if difference is using current job individual activity calendars.

How I define/change project calendar? What if compared files have different project calendars?

I am enclosing a link in reference to some of our typical CPM specs that mention/require negative float modeling as for you to understand the issues I might encounter. It is like swimming against the current. Now we have Monte Carlo but negative float is still an issue that even when not applying brand specification might still rule out Spider. On this side of the Atlantic the infatuation with negative float is very persistent.

http://www.mediafire.com/view/?2c7aprd6ch6dy7g

Of course, if negative float is hard coded as an option within Spider it might make it less difficult as a user procedure can always be questioned, can always be subject to more errors than if tested by the developer and many users.

Best Regards,

Rafael

Rafael Davila
User offline. Last seen 4 hours 32 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

Vladimir,

Thanks for the clarification about the calendar, I only got to check on using some work days instead of consecutive calendar days. I was expecting for difference to be computed using activities calendar and that if you wanted difference in consecutive calendar days a formula would do it.

I wonder if some sort of formula where you define the calendar as project/activity/specific/field could be made available and perhaps this can solve the issue in a flexible way. It would be using formulas but a script will keep it easy if using multiple formulas.

Your plans for formulas within signals looks great, it will do it for me.

I have serious issues with how most scheduling software calculate negative float by tampering with late dates of current schedule instead of using separate fields for negative float versus total float and current schedule dates versus a comparison schedule that yields the late bars of the backward computation.

Do you have any suggestions? Can it be done in Spider?

Although I have the issue with the calendars, setting aside this issue for the moment, do the bars I call negative float bars are correct and would the S-curves be correct?

Best Regards,

Rafael

Rafael,

difference in days is calculated basing on project calendar, floats use individual activity calendars. So this procedure is not quite correct. It will show negative floats but calculated using project calendar for all project activities.

For signals we usually create user fields that are calculated by formula UF=required field. Then you can create any field column name. But at the moment signals are created using certain values. In future we plan to add options to define signal values by formulas including max and min that now are not available.

Regards,

Vladimir

Rafael Davila
User offline. Last seen 4 hours 32 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

For your convenience you might download the files from the following link.

http://www.mediafire.com/?a4h2w577vn074uv,3zhgr63140ev23t

You will have to adjust the file references for the baseline and compared files locations on your computer.

Best Regards,

Rafael