To SureTrak users:
I figure it out; there is a script / (Basic Executables Type) for out-of-sequence report it is SEQUENCE.sbx It is not documented under help menu maybe because it is on Service Pack 3. I just added this and BASELINE.sbx for multiple baselines and XCONST.SBX for external constraints into the custom tool menu for easy retrieval. You can download SureTrak service pack 3 (STSrvPck3.exe) from the following link.
http://www.projectstatus.com/pub/
Still missing a report on Activities with unsatisfied and/or unachievable constraints.

To my MS Project Friends:
Good luck, don’t give up.

Maybe a filter for activities that have already started, where any predecessor were not 100% this in order not to leave out those activities that already are out of sequence but not finished yet. But seems I also cannot get those fields within the filters tool.

Perhaps in SureTrack a script would do it. SureTrak help says scripting is a piece of cake. Well, for me it is not. Wish some of our IT friends’ could give us a hand.

For MS Project I just found this reference:
http://groups.google.com/group/microsoft.public.project/browse_thread/th...

I am not giving up, what you are asking I don’t see it as a “goodie” anymore but a "must have". Yes we got to get these little rascals.

It is a good point you make about being able to differentiate between CPM and resource DRAG.

There is also the point that if you build a schedule in the manner I was suggesting, you are saying to the guys at the coalface that some of these sequences are mandatory and some not, and relying on their training / experience / supervisors to distinguish between the two, which must increase the risk of Trevor’s jackhammer moment.

In fact, you can probably add ’previous experience of planners like me’ to the list of reasons why OOS occur!

So I concede defeat here, gents. I had previously eschewed using start no earlier constraints for this type of scenario as the constraint date needs to be monitored everytime you progress the schedule, which is a lot of manual work and opportunities for error if you’re running a big schedule without resource levelling. On reflection though i can see (as in so many cases!) it’s worth putting the extra manhours into the planning to reduce the project risks.

A thought just occured to me: Would a new relationship type which constrained one activity not to occur in parallel with another be useful for this type of situation? Is that similar to this ’reverse logic link’ you describe, Steve? Possibly also useful for situations where you have mutiple tasks occuring in limited space and don’t want various trade tripping over each other? It’s really a bit a fudge to get around resource levelling I suppose, but might still fill a need. I’m struggling to get my head around any implications on calculating float, though.

Thanks to both of you, Trevor & Steve. I’m a better planner for your comments. -Chalk another one up for the PP fraternity!

I meant "Precisely, Steve".
Paul, there is always pressure to start things.
The project manager’s job is not only to make things happen but also to prevent them from happening until they are supposed to happen.
If there is no commitment to following the plan, what was the point of planning the job in the first place?
If this situation prevails, in which there is essentially no discipline, or a very relaxed attitude to it, then there will be chaos and projects will come unraveled.
If you’ve got chaos, the cause lies beyond the software.

Hi Trevor,
’OPP’ is Open Plan Professional, another planning package like P3 or MSP.

Yes you can "expect the software to know or assume that Task 3 is also a predecessor just because Task 3 is a predecessor to Task 4."
Most planning packages do this by default, with the option to turn it off if you’d rather.
I’d like to say I’m suprised that MSP doesn’t do this, but I have a low opinion of MSP.

Paul: Sorry, can’t help.

Wow! You ask a little question around here, and you get some great discussion! Paul and Gary, thanks for your explanations -- I now have a much better idea of how OOS tasks happen.

However, I have to agree with Trevor -- however it happens, it’s bad practice. As soon as an "unapproved" OOS event has occurred, (1) it needs to be identified, (2) either the plan or the work needs to be amended (either with MSP or with jackhammers!), and (3) future unapproved OOS events need to be vigorously discouraged.

One of the planners primary goals should be to, as much as possible, separate delays due to logic (Newtonian physics) from delays due to resource insufficiency. It’s usually a lot easier to correct the latter than the former, if not on the one project, then through a vigorous ABCP and post mortem process, on later ones. But only if delays caused by logic (CPM DRAG) are clearly distinguishable from those caused by resource bottlenecks (resource availability DRAG).

Gary, the scenario you describe, where the designer could do either one of two tasks, but not both, should be handled other than by inputting a pred-succ relationship. In the old MicroPlanner, there was a function called a "reverse logic link," where shorter path to a merge point would be scheduled earlier, and this might have worked for the case you describe. But in other packages (e.g., MSP), I think the situation is better addressed by using a no-earlier-than scheduling constraint for one of the design tasks rather than using a logic relationship that’s not based on work logic.

Not that scheduling constraints can’t also be problematic -- it’s just that their DRAG can be easily identified, whereas a resource bottleneck passing as a logic delay muddies the waters and makes it impossible to see that, by paying an extra $20,000 for an additional designer, the CP could be pulled in by four weeks at a value of $50,000 per week.

In my book, I reference a conversation that took place about 20 years ago, with the head of the program office for the F-16 in Fort Worth, Texas. "I can’t get my engineers to work to the CP! They’ll plan to it, but then either they do whatever tasks they feel like doing, or whatever gives the most earned value. The result is that the most amazing things happen! In Feb and March, the plane is on schedule and on budget -- then in April, it’s suddenly ten months late, and $15 million over budget! And it’s all because of OOS work, which then stops you from ever knowing where you really are!"

Fraternally in PM,

Steve D.

Paul:

For your convinience I uploaded my test file which includes the unfinished macro. If you wish you can download it from the following link.

http://rapidshare.com/files/227505472/OOS-03.mpp.html

Will be available with the following limitations:
Your file has been saved and can now be downloaded 10 times. It will be deleted after 90 days without download. If you would like to enable more people to download your file, please transfer it to a free Collector’s Account or a Premium Account.

Still pending modifications to macro to account for SS, SF, FS and FF relationships all with lag.

So I guess we agree to disagree.
If "many logical links are there in the schedule not becuase A MUST happen before B..." then they are wrong in the first place, and they are proven to be wrong because B happens before A.
If there is no reason that B cannot start on day 1, then it should be scheduled to start on day 1.
CPM is about finding out what is possible, it goes to feasibility.
The task’s start date is always the earliest start date, meaning the earliest possible start date.
The date that it actually starts can be any date after that as long as it does not soak up all of the total float on the path.
Have you never seen a colossal screw up caused by someone who decided to take it into his head to know better how the job should be done and chose to ignore the plan?
Do you know how much time and money it costs to jackhammer out 16 tonnes of concrete?

Mr. Dua, as others, I do remember Micro Planner software, specifically Micro Planner for the Mac., while everyone else was hooked on the DOS version of P3. I just loved it, it was available for very good price, had the ability to work either activity on arrow or activity on node networks, and many other goodies. Yes, it would be nice if we all have the option available in Micro Planner to schedule the successor dependant task at the Earliest of the earliest starts.

I believe the more of these tools to model real life situation the better. I don’t buy the statements of some “Gurus” (http://www.pmicos.org/fse.asp) that these tools are invalid because they create confusion. Sorry “ladies” we need tools that mimic real life situations.

MS Project is not my software of choice, but Paul is using MS Project. Paul asked for a way to be alerted. If we all agree that there is no way to predict the “unforeseen”, probably one of you can provide us with a practical mean to be warned of out of sequence. Also, what about a mean to filter out those activities that are out-of-sequence within the means of our software? I understand some software packages do provide for this, I am still looking in Suretrack for a direct method that is neither cumbersome nor time consuming. Maybe I can learn something from the MS Project solution I could adapt to my software.