Dear Planners,

An example to the question of the topic:

Process Engineering
Activity A Interface Matrix Study
Activity B System X P&ID
Activity C Line List

You can start working on P&ID’s without waiting the interface study but you cannot finish the work without completing it.

You shall start studying the Line List with the P&ID’s. (If you are certain about all lines handled you can issue a Line List before P&ID and leave the details to the ISOS and other data sheets, Specs etc.)

A – FF – B
B – SS – C


Just an example when I brainstorm there will be another solution for sure. I am also a supporter of FS links but the life of a planner is not just easy to handle every situation with FS’s (Especially information gathering and time constraints are in real).

Regards,

Safak VURAL

Rafael

Thanks for that

I especially like your idea of all learning centres having a couple of copies of Micro Planner.

In fact in Australia and new Zealand that is the case with some universities who teach Arrow and Precedence Diagrams.

It is only our friends! in the US who have trouble with ADM

We do in fact provide the Universities with cheap version of X-Pert, provided it is not used commercially.

Raf

Folks

According to Max Widemans Glossary a Hammock is an

Activity, joining two specified points, that span two or more activities.
Note 1: Its duration is initially unspecified and is only determined by the durations of the specified activities.
Note 2: Hammocks are usually used to collect time-dependent information, e.g. overheads. [D04582]

According to what we described in ICT back in 1964 when we developed Hammocks in ICT 1500 PERT

Hammock Activity

A Hammock is a special link that stretches from the start of one task to the finish of another. Unlike other links, it calculats its duration by figuring out the elapsed time between the two tasks. In other words, you do not allocate a duration to a hammock -it assigns its own. Furthermore, a Hammock link can carry all the information that you would enter for a task, such as resources and labels.
The primary function of a Hammock is to apply fixed resources or costs over a particular phase of the project -the longer the Hammock stretches, the more it costs.
In addition, Hammocks can be used to summarize parts of the network as single bars on a Gantt chart. If these are placed strategically through the network, they can be used to report the broad outlines of the project -minus the detail.

When you are describing "Hammock" is this what you mean?


Every time a new piece of software comes out existing terminology seems to take on new meanings.

Raf

Mike,
you wrote: No - the buffer task starts when the handover period begins - FS - and ends when the new chap starts again FS.

Do you mean that buffer task is hammock and the link is from hammock activity to some regular activity?

Hammocks are linked different ways in different software. In Spider Project links are TO hammocks because their start and finish dates depend on other tasks and not vice versa.

In any case people shall do their work until the finish event happens. Why buffer and what does it mean?

Best Regards,
Vladimir

"Soft link" = Where there is a limited resource trade on a number of work fronts I use a FS link between the locations. They are called soft because the link appears and disappears with timing and resource availability.

If you increase resource availability the link might no longer exists, then if you don’t remove the relationship link (call it soft link or resource link if you will) you will get the wrong schedule. The schedule cannot be unleveled under the presence of these "soft links".

The term is everywhere in the US CPM literature, and some of our specs prohibit the use of soft links.

Hi Vladimir

"create a buffer task that will be in place for as long as is necessary."

"Yes and it is necessary until the next task starts - still SF link."

No - the buffer task starts when the handover period begins - FS - and ends when the new chap starts again FS.

BTW Powerproject does work volume lead lags too - never used them.

Best regards

Mike Testro

Hi Suraj,
we still don’t recommend to use SF links.
Only few occasions justify them and these accasions are rare.

The procuremet and installation activities shall be linked by FS link with the positive Lag but the procurement activity shall be defined as ALAP (if you don’t use MS Project!).

SF means that procurement can finish some time earlier that the installation (negative lag time) but may be delayed without delaying the installation!!!

Best Regards,
Vladimir

Only ever found one use for SF relationship - shift handover/equipment replacement.
E.g. the ’old’ chap cannot leave until the ’new’ chap has started, or the old equipment cannot be removed until the new equipment is installed and working/commissioned.

That could be valid as well considering a bulk foundation on grade where you can strike the side formworks in a few hours as long as the concrete has set and not be damaged.

Concrete curing is not necessarily a fixed time. It is to be specified.

How about a situation as follows...
"Pour(Bulk) Concrete" with a SS to "Curing Time"?
Or even a FF with lag?
Could you say that was valid?

Folks

At the 2006 PMI COS conference a paper entitled “Scheduling 101: A “Behind – the – Scenes” Look at Basic Schedule Calculation was presented by Scott W Cramer, James L Jenkins, and Joseph J Orczyk¹. In the paper they present a “Comparison between AON and PDM” (page 8), in which they make a case for using Precedence Diagram Method for overlapping tasks using three basic construction tasks. They then demonstrate the process of overlapping the three PDM tasks and how the critical path computations are calculated.

In their example of using Start to Start and Finish to Finish links to simplify the network, they clearly demonstrate a major flaw in the PDM calculations. This flaw is known to many planners and schedulers as “Lag Drag”, where the Earliest Start of the task is dragged to the right, because the software used subtracts the duration from the Earliest Finish Date, Jenkins, Kramer and Orczyk state “….. the ES value is then calculated by EF minus Duration

Lag Drag is where the Earliest Start of a Task is dragged to the right due to a longer task progressively feeding a shorter task and SS and FF links have been used to define this overlapped work. This phenomena is due to the way PDM computations are calculated, .i.e. ES = EF – duration rather than the normal ES calculation as defined by Henderson 1998 ² which says that “The earliest start time for each activity (i,j) is equal to the earliest possible time for the preceding event E(i):” in other words

ES = PEE

Using an arrow diagram to define overlapping (progressive feed) activities does not suffer from the lag drag problem provided the software programme follows the ladder convention

I have a paper which I presented at PMICOS earlier this year which defines the term Progressive Feed and the definition of a Ladder and demonstrate as well the appropriate calculations to prevent lag drag.

As I was one of the three people at ICT back in 1964 who created Ladders, I think I have a fair degree of understanding of what the Planner is meant to be doing when using the Ladder Facility, which the PDM tried (and failed) to mimic with SS and FF.

I cannot see how one would have any case of using a start to start after a finish to finish


My paper is 14 pages,which is bit o the long side to include here. If you would like a copy e-mail me at raf@microplanning.com.au

You may be interested to know that CIOB in the UK are looking at the use of Ladders in complex project planning

BTW is there a way for me to insert a word file into a Planning Planet response

Raf

Hi Shahul

When has anyone ever used a SF link between tasks?

You would be building it backwards.

Best regards

Mike Testro

Gary,

Driving can kill you or you can kill others, even both. Therefore driving is to be banned in the UK. If not at least you should be banned, you are a wreck less driver. Watch out I am worst, especially if driving in one of the few countries that drive on the wrong side of the road, crazy people. Your license is hereby suspended for 3 years.

FF followed by SS seems like a bad idea. Precisely this is the kind of rules that all CPM software should have embedded so that as soon as the software detect this a dialog box appears warning you about the risks and unless you waive it within the dialog box the software will not let you apply the link.

I always avoid using SF and FF relationships, tying a predecessor to the finish of a successor are kind of awkward, but “never say never”. My exception to this rule is when I try to mimic “laddering” a special type of activity created by the British, my approach is not the same and still within limits can cause some awkward logic results. Here I use SS (no problem) and FF (slippery when wet) between the two activities, both the same duration and with equal lags. True laddering as per Micro Planner is a safer and meaningful but my software lacks this functionality.

I try to limit myself to the use of SS and FS relationships, the predecessor tied only to the start of the successor.

With great pain in my heart I got to say that for the specific case you are mentioning I would follow Mike Testro’s Bottom up procedure and avoid the FF link.

Best regards,
Rafael

Gary

could you please explain by a practical example where such relationships may be valid as appose to FS?

regards

Shahzad

To Mike: No!

To Nigel: Successor activity usually shall be of hammock type.

Best Regards,
Vladimir

Hi Vladimir

So if the new chap gets run over by a bus on the way to work the old chap waits for ever?

You are adding resource management to a linkage discussion.

Best regards

Mike Testro

This is the logic of this link. Old chap cannot leave until the start of the following activity will happen.

If old chap is air dispatcher and will leave before another dispatcher came, a lot of people will be endangered.

Proper links shall work with resource management, in other case it is poor logic.

Best Regards,
Vladimir

Mike,
you wrote:
create a buffer task that will be in place for as long as is necessary.

Yes and it is necessary until the next task starts - still SF link.

Best Regards,
Vladimir

Hello,

i am in the begining of my career and obviously do not know the subject as much as you do but as far as i understand, when we are making a schedule we cater all resources and then activites are put in the form of ghantt chart.

The problem of the person responsible for the 2nd activity is not to be considered when defining a schedule or while making a logic.

I may be corrected if i have got it wrong.

Regards,

Basir Ahmed

Guys!!
I am not an expert but SF can come handy when a Client is on your head for a Master Program (assume you are a PMC and on board a day before- I am sure this will sound familiar to many)
So what I do (with all due respect to right planning practices):-

I make a construction schedule (may be level 3).
I have to link my procurements to my deliveries and execution.
I link the the activity (start procurement process of Boilers under Procurement WBS) to its predecessor (Installation and comissioning of boilers under MEP works WBS) with SF relationship+ some buffer(-ve lag) to account for procurement delays.
This gives me a fair presentation of when the procurement of Boilers should start to get the work executed in time as planned.
BR,
Suraj

Hi Mike,
you wrote:
BTW Powerproject does work volume lead lags too - never used them.

Sorry, I did not notice it but I did not play with PP long enough.
Please explain how to set in PP SS link with 500 meters lag.

Best Regards,
Vladimir

Hi Rafael

Now I understand.

This is starting up as a new version of lets challenge spider.

There is nobody more enthusiastic as a new convert.

Lets get back to KISS shall we.

Best Regards

Mike Testro

I know definitions of hammocks and LOE activities.
Having too many activity types is confusing to users.

In Spider Project hammock is an activity which start and finish are defined by other activities or imposed dates.
So hammock duration is flexible and it lasts from one event in the schedule to another (not necessary linking start of one and finish of another activity).

But we permit to assign resources to hammocks, without changing activity type to LOE.
And we permit to create several start and finish conditions for hammocks. It starts if one of initial events is happened and ends when all finish conditions are satisfied.

I don’t know what is WBS activity.

Spider Project is on the market since 1993. It is widely used in Russia for more than 15 years. I would not call it as the new software.

I know Critical Chain theory. And I know what buffers mean and how are used in this theory. PowerProject is not Critical Chain software and PP buffers are not the same as buffers in CCPM.

I think that it would be interesting to discuss buffers and Critical Chain approaches but not in this discussion. I think that our approach to creating and usage of buffers is much more practical. It may be an interesting discussion.

Look at http://www.spiderproject.ru/library/eng/Programs%20in%20Russia.ppt for more information on our approaches.

Best Regards,
Vladimir