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.

P6 & Critical Chain

8 replies [Last post]
Paul Burton
User offline. Last seen 6 years 7 weeks ago. Offline
Joined: 16 Aug 2007
Posts: 6
Groups: None

Hi,

Does anyone know of any tools on the market that interface with Primavera P6 to support multi project critical chain implementation?

Thanks

Replies

Rafael Davila
User offline. Last seen 13 hours 30 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

The following references might also be of interest.

5. Conclusions and suggestions for further research

Using the 50% rule for buffer sizing may lead to a serious overestimation of the required project buffer size. The 2σ buffer size assumption does not hold. Keeping the critical  chain  activities  in  series  is  harmful  to  the  final  project  make span.  The WIP impact of the scheduling mechanism used for scheduling the gating tasks is negligible. Regularly updating the baseline schedule has a strong beneficiary  impact  on  the  final  project  duration  and  provides  better  intermediate  estimates  of  the  final  project make span.

While I am not an enthusiast of CCPM I believe it is the end user, the one responsible for the means and methods, who shall decide.  For those advocates of CCPM the following article might be of interest.

Critical Chain project management and NEC3

CCPM is a simplified methodology for buffer management, good for when you cannot make good plans that make use of much needed buffers.  It brings nothing new to Good Resource Critical Path Scheduling other than a reminder that better planning makes use of buffer management.

Specific CCPM methodology and software shall not be mandated.  In any case the use of project buffer such as the NEC3 requirement makes a better approach.  Just leave the CCPM methodology as an option.

Same as there is Bad Multitasking there is Bad Scheduling.

Paul Burton
User offline. Last seen 6 years 7 weeks ago. Offline
Joined: 16 Aug 2007
Posts: 6
Groups: None

1.2&3 : Do not require much knowledge of CPM to understand that changes in critical chain are more frequent than not.

 - Less often than critical path in my experience. The point is that "does not change" has never been claimed.

4. : Amount to be removed? Can be 50% as proposed by Goldrat but why not 75% or 100%.  If the amount to be removed is arbitrary then the formula is arbitrary.

 - Because, to a reasonable level, the activity is only 50% likly to be achieved by the new finish date. Buffer is expected to be used, individual activity end dates are not tracked. Whatever duration you remove, the formula is not arbitrary. It is a relatively common method for statistical toerance analysis.

5. : Following your reasoning for the crane there is no such thing as multitasking, it's completing one task, then doing the next. I wonder why prohibiting something that never happens.

 - I think a gap in understanding. Really the right term here is "bad multitasking". This is where an activity is started but not finished, then another activity is started. When the first activity is re-started there is a delay where it has to be picked up again, what was done is reviewed, etc, etc. A crane can't do that (unless you're employing some anti-grav tech I'm not aware of). This may not be the regular definition of multi-tasking, but it is what is meant in this context. NB: You originally talked about one activity per day which is not a requirement of the technique.

6. : That scheduling is not done in manufacturing is new to me, I've no experience of manufacturing. TOC originates in manufacturing where it is feasible to protect Critical Chain.  TOC can be effective in manufacturing, but not on all environments.

 - Work is often scheduled, but this is not project planning/project scheduling - reference definition of "project".

7. : To prevent waste due to waiting inventory - in other industries waiting inventory is not the norm, ALAP shall not be applied to all activities. In construction contracts indiscriminate application of ALAP is considered float sequestration.

 - Build a wall and leave it in the elements for six months, while it waits for something else to happen, it will, likely, degrade. This is the definition of inventory causing causing systemic issues.

8: In over 40 years doing construction work at risk I do not recall a single job where liquidated damages were applied. I do not recall a single project without change orders all paid by courtesy of the client.  I do recall a couple of construction claims awarded by the court in favor of the contractor.  Over 90% of the delays were due to owner and those due to contractor were taken care easily before impact. No matter how bad contractor might be the competition is such that the markup is minimal, there is not much room for contractor to fail in planning, the failure is on others, this has nothing to do with the scheduling methodologies.  This will not change until industry start looking at the true source of overruns and those looking to take advantage of it stop hard selling promises in the air.

 - My experience is very different, I'm afraid.

Rafael Davila
User offline. Last seen 13 hours 30 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

The following references might also be of interest:

CCPM - Inserting buffers in a resource-constrained schedule

CCPM - The problem with resource conflicts

CCPM - Sizing project and feeding buffers

Evaluating performance of CCPM

Currently, there is no standardized approach available to cope with these resource conflicts generated by the insertion of the buffers. Resizing buffers to avoid resource conflicts can be considered as a last and probably the most simple and pragmatic technique to resolve resource conflicts. As an example, the buffer sizes obtained by the cut and paste method as displayed in figure 1 do not lead to resource conflicts.

It should be noted that all these methods to resolve resource conflicts might result in a buffered baseline schedule where resources are not over allocated, but the risk might be that the resulting buffering approach is no longer capable of absorbing potential delays in the chains feeding into these buffers. Therefore, these methods must be applied with utmost care. The sizes of the buffers should always be a reflection of the uncertainty of the activity duration in the chain leading to the buffer and the buffered schedule should therefore be constructed so that it is capable of coping with potential delays during the progress of the project.  Simply resolving resource conflicts might put that objective into danger.

Rafael Davila
User offline. Last seen 13 hours 30 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

1.2&3 : Do not require much knowledge of CPM to understand that changes in critical chain are more frequent than not.

4. : Amount to be removed? Can be 50% as proposed by Goldrat but why not 75% or 100%.  If the amount to be removed is arbitrary then the formula is arbitrary.

5. : Following your reasoning for the crane there is no such thing as multitasking, it's completing one task, then doing the next. I wonder why prohibiting something that never happens.

6. : That scheduling is not done in manufacturing is new to me, I've no experience of manufacturing. TOC originates in manufacturing where it is feasible to protect Critical Chain.  TOC can be effective in manufacturing, but not on all environments.

7. : To prevent waste due to waiting inventory - in other industries waiting inventory is not the norm, ALAP shall not be applied to all activities. In construction contracts indiscriminate application of ALAP is considered float sequestration.

8.  : In over 40 years doing construction work at risk I do not recall a single job where liquidated damages were applied. I do not recall a single project without change orders all paid by courtesy of the client.  I do recall a couple of construction claims awarded by the court in favor of the contractor.  Over 90% of the delays were due to owner and those due to contractor were taken care easily before impact. No matter how bad contractor might be the competition is such that the markup is minimal, there is not much room for contractor to fail in planning, the failure is on others, this has nothing to do with the scheduling methodologies.  This will not change until industry start looking at the true source of overruns and those looking to take advantage of it stop hard selling promises in the air.

Paul Burton
User offline. Last seen 6 years 7 weeks ago. Offline
Joined: 16 Aug 2007
Posts: 6
Groups: None

1, 2, and 3 are the kind of strawmen I'm talking about.

4.  Square root of the sum of the squares of the amount removed from each task estimate.

5.  Your crane is not multitasking, though. It's completing one task, then doing the next.

6.  No-one does project planning in manufacturing because, by its very nature, manufacture is a serial and repeated set of processes. Theory of constraints orginates in manufacturing and this work was adapted for project planning. It can be effective in research, design, and development environments - I've no experience of construction.

7.  "CCPM advocates starting tasks as late as possible to prevent re-work" - to prevent waste due to waiting inventory.

8. "These problems do not require the introduction of different techniques, such as CCPM, to get them under control. These behaviors can be easily introduced in the traditional project management environment" - and yet the vast majority of projects are late, de-scoped, or worse. Often despite the huge efforts of highly skilled and committed people.

Rafael Davila
User offline. Last seen 13 hours 30 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

The internet is full of papers challenging Critical Chain procedures.

Mostly CCPM is criticized because of:

1.  The assumption that Critical Chain will never vary because it can always be protected.

2.  The assumption that Critical Resources will always remain critical and non-critical will always remain non-critical.

3.   The assumption that Feeding Buffers will always remain feeding into a static critical chain.

4.   Buffers are arbitrary and no methodology is provided on how to calculate necessary buffers.

5.   CCPM recommends that multitasking be eliminated when at times it is necessary, not always a burden. Take for example the case of a tower crane that within a single day must be shared among many activities bur must be resource leveled as not to be assigned more activities it can handle in a single day.

6.   CCPM was developed for manufacturing environment which is different from construction.

7.   CCPM advocates starting tasks as late as possible to prevent re-work.

8.   These problems do not require the introduction of different techniques, such as CCPM, to get them under control. These behaviors can be easily introduced in the traditional project management environment.

The following papers discuss these and a few more in some detail.

Resource Critical Path Approach to Project Management by Vladimir Liberzon

Five Behaviors That Can Reduce Schedule Risk

An analysis of Critical Chain concepts

Paul Burton
User offline. Last seen 6 years 7 weeks ago. Offline
Joined: 16 Aug 2007
Posts: 6
Groups: None

Thanks I'm looking into spider project.

Rather a dodgy old paper that. Knocks down plenty of strawmen and certain sections don't even make any sense. There must be better critiques than that one.

 

 

 

Rafael Davila
User offline. Last seen 13 hours 30 min ago. Offline
Joined: 1 Mar 2004
Posts: 5229

You can try Aurora, it can be used with P6 or standalone.

https://www.stottlerhenke.com/products/aurora-ccpm/

Aurora is perhaps the best American CPM scheduling software, but pricey.  Same as Spider Project it is capable of dealing with spatial resources, optimization algorithms, and a many other things P6 cannot do.

Out of the box I got better result, for Aurora sample project of 8 activities with Spider Project but the capabilities of Aurora shames P6 and MSP.  I got the known optimal solution.  Still resource leveling optimization is very complex and I do not pretend to say Spider or Aurora is better than the other, a fair comparison would require many test projects.

https://www.stottlerhenke.com/wp-content/uploads/2011/09/2009-10-01_Aurora_WhitePaper_Turnaround.pdf

I believe CCPM to be flawed, I do not expect CCPM advocates to agree with me. It is wrong to assume Critical Chain and location of Feeding Buffers will not change.

http://csbweb01.uncw.edu/people/rosenl/classes/OPS100/A%20Critical%20Look%20at%20Critical%20Chain%20Project%20Management.pdf

Good Luck