Productivity, Performance, & Earned Value

 View Only
Back to discussions
Expand all | Collapse all

Contingency Drawdown RP

  • 1.  Contingency Drawdown RP

    Posted 12-08-2021 08:35 AM
    Hi, I am thinking of developing a scoping document for an RP on contingency drawdown. It would propose two alternative approaches; stochastic vs quasi-deterministic, or mechanistic. My approach is stochastic, but many if not most use mechanistic. Stochastic accepts that 1) most risk are systemic and therefore cannot be effectively linked to the WBS or activities, and 2) most critical risks in a register or QRA do NOT happen so using the register for expenditure planning is largely futile. My approach uses empirical research of contingency spend patterns (correlated with labor) and principle of frequent re-quantification to address dynamic risk emergence and change. Mechanistic on the other hand assumes that all risks can be attached to the WBS, that the risk register items will all occur (considering probability) and does not hinge on frequent re-quantification. The mechanistic approach was most recently presented by Stroemich and Dissanayake in 2018 Transactions. I have not written a paper yet on stochastic, but have my own procedures. Is there any interest in this? Ideas? Suggested references?
    PS: there is also a recent paper by Kozak and Norfleet; it is focused more on tracking use of funds rather than developing the original curve. It appears to presume the mechanistic, risk register approach. It also provides good US government insight (which leans to mechanistic) as well as schedule. There is a 2015 paper by White that has some interesting views on risk types, but also appears to use the mechanistic risk register method (or equivalent via a CPM model with all risks included).

    ------------------------------
    John Hollmann CCP CEP DRMP
    Owner
    Validation Estimating LLC
    Leesburg VA
    1 (703) 945-5483
    ------------------------------


  • 2.  RE: Contingency Drawdown RP

    Posted 12-09-2021 08:30 AM
    Risks not necessarily are linked to activities. Example: uncertainty in resource availability,
    I don't understand why mechanistic approach cannot be supported by frequent re-quantification.
    Frequent re-quantification is required for risk and project management in any case.


    ------------------------------
    Vladimir Liberzon
    General Manager
    Spider Project Team
    Moscow
    ------------------------------

    Original Message


  • 3.  RE: Contingency Drawdown RP

    Posted 12-09-2021 08:51 AM
    Vladimir, no reason mechanistic cannot also involve re-quantification. I agree frequent re-QRA should be required. And if you do so, the "curve" becomes somewhat of a non-issue. Risks and behaviors never stand still. However, I find that mechanistic-oriented companies tend to lean on the risk register heavily and believe that by maintaining it, and linking to it, regular re-QRA is not needed (i.e., the register IS essentially their QRA tool). In my experience, this does not work. Second, mechanistic-oriented approaches often tend to use risk-driven (tying all risks to activities) CPM modeling which is a challenge to maintain/re-run frequently. More often then not, I find an external expert ran the model for the owner originally and calling them back regularly is not practical. So mid-stream QRA ends up not being done until its too late (but I have seen a mega-project running their CPM risk model every month; so certainly possible if tools and resources are on hand). These are kinds of strenghts/weaknesses, pluses/minuses we may want to touch on.

    ------------------------------
    John Hollmann CCP CEP DRMP
    Owner
    Validation Estimating LLC
    Leesburg VA
    1 (703) 945-5483
    ------------------------------

    Original Message


  • 4.  RE: Contingency Drawdown RP

    Posted 12-09-2021 08:36 AM
    I am informed Lance Stephenson has contingency management RP on his plate, so a scoping doc is forthcoming. But if anyone has thoughts on scope, post them here so they get consideration. In any case, the scoping doc would be shared in Community to make sure its on right track.  Some issues in offline discussion already brought up are:
    -principles (e.g., my stochastic vs mechanistic comments)
    -whether to cover just contingency per se, or all risk funds (contingency, management reserves, escalation, currency)?
    -cost only, or schedule too?
    -owner vs contractor vs government (e.g., regulated)
    -integrating this with change management RP(s)
    -integrating this with QRA RPs
    -programs and portfolios (if reserve management is addressed)
    -others

    ------------------------------
    John Hollmann CCP CEP DRMP
    Owner
    Validation Estimating LLC
    Leesburg VA
    1 (703) 945-5483
    ------------------------------

    Original Message


  • 5.  RE: Contingency Drawdown RP

    Posted 12-09-2021 11:25 AM
    My answers:
    - stochastic
    - contingency
    - schedule too
    - owners and contractors can manage different risks
    - integrating with change management and not with QRA RP that does not cover everything
    - programs and portfolios

    Risk simulation must include risks that are not associated with specific activities (resource availability, calendars, etc.) or lead to changes of project scope and replacement of project activities with the different sets. Besides it must model corrective actions people apply when meeting project targets is endangered.
    -

    ------------------------------
    Vladimir Liberzon
    General Manager
    Spider Project Team
    Moscow
    ------------------------------

    Original Message


  • 6.  RE: Contingency Drawdown RP

    Posted 12-23-2021 04:34 AM
    Perhaps, one very important aspect of Contingency drawdown is the trigger point, the point at which the risk becomes a reality.  The project manager and the risk manager have to decide (judgment call) when to trigger the contingency drawdown such that the project can get the best value of the contingency by drawing at the right time. My suggestion is to touch on the importance of time in both the identification, duration, and trigger-point of risk, that is, risk and contingency, as a function of time.

    Rufran

    ------------------------------
    Rufran Frago P. Eng. PMP CCP PMI-RMP
    Founder/Business and Project Consultant
    PM Solution Pro |RBM&S Inc.
    Calgary AB
    ------------------------------

    Original Message


  • 7.  RE: Contingency Drawdown RP

    Posted 12-09-2021 05:27 PM
    All, I am getting private posts with ideas. Its interesting, but not very helpful (especially if I am not ultimate lead RP author). Much preferred that Community members post in the Community on the topic so others can chime in. And it keeps everything in one place for reference later. All input is good. If for some reason you cannot post publicly, let me know, and I will copy the idea over here anonymously. Thanks!

    ------------------------------
    John Hollmann CCP CEP DRMP
    Owner
    Validation Estimating LLC
    Leesburg VA
    1 (703) 945-5483
    ------------------------------

    Original Message


  • 8.  RE: Contingency Drawdown RP

    Posted 12-13-2021 07:38 AM

    Hi John,

    In Fluor we have not gotten stuck on the statistical terms although I get from your message that it ties into the two (or three if we include schedule contingency) mechanisms that we have for managing contingency.

    We make a clear distinction between our "estimating" contingency and event driven contingency. The former is based purely on uncertainties in the estimate and is established using a Monte Carlo Simulation and selecting a P level that our company is comfortable with. This contingency is managed using a rundown (and is expected to be spent and thus run down to zero when the job is complete). I distinguish between rundown (showing your "forecast to complete" contingency against an expected rundown curve) and a drawdown (a curve that forces you to reduce your "forecast to complete" based on achieving specified milestones). I am not sure if this is or could be an accepted definition for rundown vs drawdown? Our preference is a rundown (using this definition) since it helps to stabilize the bottom-line forecast and having a better 'justification' for forecast changes.

    We have a second value that is based only on identified risks of events that may or may not happen, and by applying a likelihood of occurrence, an even driven risk contingency is established (again with a Monte Carlo Simulation). This is not reported against a rundown but re-evaluated on at least a quarterly basis. Events that do occur (or are likely to occur) are removed from this simulation and included in our (mitigated) cost forecast. This contingency may have a residual value when job is completed due to remaining risk of liabilities - when all events either occurred or will not happen at all, this contingency value is zero.

    Best regards
    Anton

     

    Anton van der Steege | FLUOR | Fellow - Global Cost Lead | CCP | anton.van.der.steege@fluor.com

    www.fluor.com | Facebook | Twitter | LinkedIn | YouTube

     

    ------------------------------------------------------------
    The information transmitted is intended only for the person
    or entity to which it is addressed and may contain
    proprietary, business-confidential and/or privileged material. 
    If you are not the intended recipient of this message you are
    hereby notified that any use, review, retransmission, dissemination,
    distribution, reproduction or any action taken in reliance upon
    this message is prohibited. If you received this in error, please
    contact the sender and delete the material from any and all
    computers and other devices. 

    Any views expressed in this message are those of the individual
    sender and may not necessarily reflect the views of the company. 
    ------------------------------------------------------------



    Original Message


  • 9.  RE: Contingency Drawdown RP

    Posted 12-13-2021 09:45 AM
    Thanks Anton, I am on the hook for RP RM-34 that quantifies contingency using a hybrid Ranging + Expected Value method which likely compares to what Fluor is doing. Its quite common at EPCs. The rundown vs. drawdown distinction is useful; both terms will need to be added to 10S-90 (perhaps some alternate definitions; another word our industry uses in various ways). Your quarterly re-quantification is good; it assures that one is not caught aware by relying on quirks in mechanistic approaches. Supports my view that risks on major projects are just too volatile/dynamic to rely on a more or less static "plan" such as risk registers lead to.

    ------------------------------
    John Hollmann CCP CEP DRMP
    Owner
    Validation Estimating LLC
    Leesburg VA
    1 (703) 945-5483
    ------------------------------

    Original Message


  • 10.  RE: Contingency Drawdown RP

    Posted 12-14-2021 09:50 AM
    It would be great to have a distinct recommended methodology for contingency drawdown for each of the following methodologies that are likely the most used by the community:
      - parametric + EV hybrid
      - CPM using risk drivers
      - hybrid ranging + EV (in progress)
      - hybrid parametric + CPM (in progress?)
      - EV only

    Also, I'm curious if anyone has managed schedule contingency in a similar way cost contingency is managed.   From my understanding, CPM based methodologies would be the only way to tie schedule contingency to specific milestones or activities.

    Qs for ranging + EV: If ranging for only estimate uncertainties, will this be assuming the project is not complex and there are no or little systemic risks based on the revised/new range estimating RP that will be coming out soon?  Also, how would the overall project contingency value compare with the managed contingency if MCS is run separately for ranging and EV and the project does not select a P value close to the mean?

    ------------------------------
    Andre Macatangay
    ------------------------------

    Original Message


  • 11.  RE: Contingency Drawdown RP

    Posted 12-14-2021 04:12 PM
    Since 90-s we use an approach to project performance analysis that is called Success Driven Project Management (SDPM),
    It is based on analysis of contingency reserves drawdown based on trends of probabilities to meet project targets,
    Negative success probability trends show that project buffers (contingency reserves) were consumed faster than expected and corrective actions must be considered.
    Look at http://www.spiderproject.com/images/img/pdf/Success%20Driven%20Project%20Management/Methods%20and%20Tools%20of%20Success%20Driven%20Project%20Management.pdf and let me know if you will have any questions.

    ------------------------------
    Vladimir Liberzon
    General Manager
    Spider Project Team
    Moscow
    ------------------------------

    Original Message


  • 12.  RE: Contingency Drawdown RP

    Posted 12-15-2021 09:57 AM
    Hi John,

    Somewhat tangential to the proposed RP's stated purpose, I would like to suggest as a means of communication between contractors and owners the proposed RP address the concept of "liens on contingency," risks that the project has not formally realized but may have already impacted the project.  An owner would expect to see a running list of liens on contingency in routine (monthly) reports.  One or more related change requests would follow an addition to this list.  This helps the owner better prepare for schedule and cost contingency requests.

    ------------------------------
    Ivan Graff CCP
    Program Analyst
    U. S. Department of Energy
    Germantown MD
    1 (301) 903-8437
    ------------------------------

    Original Message


  • 13.  RE: Contingency Drawdown RP

    Posted 12-16-2021 08:20 AM
    Ivan, not tangential at all. In a similar vein, when I used to do benchmarking, I would ask to see the project change log. If the list was long and most of the items were labeled "pending", I knew the project was in (or getting into) trouble. A good control metric in that regard is to track "average days to disposition" or percent pending, and reports highlighting laggards. Requests/notices should not fester for more than a week or two. Usually the situation was worst when the PM did not own contingency (i.e., contingency management by committee or held by higher level authority). In any case, this lag/lien effect would play into contingency management.

    ------------------------------
    John Hollmann CCP CEP DRMP
    Owner
    Validation Estimating LLC
    Leesburg VA
    1 (703) 945-5483
    ------------------------------

    Original Message


  • 14.  RE: Contingency Drawdown RP

    Posted 12-24-2021 09:22 AM

    The rundown vs. drawdown discussed earlier is a very practical method.  Also, as discussed involving the Project Manager is key.  However, the Project Manager often has extreme pressure to maintain the budget, putting pressure to be optimistic and consume contingency.   A process that requires the PM to mechanically monitor the contingency, balanced with input from a neutral third party (Accounting), and appropriate oversite can balance the process. 

    Management of contingency is more than math.  A balanced process allowing the decision maker to effectively manage the process after receiving input from operations and accounting can be effective.     



    ------------------------------
    William Corn
    Managing Partner
    GMG Partners, LLC
    Murrieta CA
    (808) 216-0410
    ------------------------------

    Original Message


Related Content