ABET Civil Engineering Program Criteria Draft

Expand all | Collapse all

_5. ... apply probability and statistics to address uncertainty and risk ...

  • 1.  _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted 04-28-2021 04:20 PM

    CURRENT CRITERIA

    PROPOSED CRITERIA

    RATIONALE FOR CHANGE

    apply probability and statistics to address uncertainty;

     

    apply probability and statistics to address uncertainty and risk relevant to civil engineering;

    • The CEBOK3 explicitly requires the ability to apply probability and statistics "to determine risk relevant to civil engineering."


  • 2.  RE: _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted 05-30-2021 05:43 PM
    5. "to determine risk relevant to civil engineering."

    CE work that involves relevant and significant phase-risk includes, and is not limited to:
    a. Planning Phase.

    b. Design Phase.

    c. Construction Phase.

    d. Operations Phase.

    Suggest that these four (4) phases, along with types of risk,  be specifically identified.
    NOT as limiting factors but as common examples.

    Stay Healthy!
    Cheers,
    Bill

    ------------------------------
    William M. Hayden Jr., Ph.D., P.E., CMQ/OE, F.ASCE
    Buffalo, N.Y.

    "It is never too late to be what you might have been." -- George Eliot 1819 - 1880
    ------------------------------



  • 3.  RE: _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted 06-02-2021 07:16 PM
    Dr. Hayden,
    Thanks for your suggestion.  This level of detail would almost certainly be too high for inclusion in accreditation criteria--but it would be entirely appropriate for our ASCE Commentary to the criteria, which provides guidance to programs and evaluators on how to interpret and satisfy the criteria.  The Task Committee will certainly consider your suggestions for this purpose.

    ------------------------------
    Stephen Ressler Ph.D., P.E., Dist.M.ASCE
    Professor Emeritus
    Bethlehem PA
    ------------------------------



  • 4.  RE: _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted 06-03-2021 06:34 AM
    This proposed criteria does not exclusively call for, require, or limit uncertainty to computation, correct?  The terms probability and statistics seem to be strongly associated with computation/calculation, numbers.   Uncertainty and risk as a topic is not entirely a computation/calculation.  I seem to recall discussion of this in some CEPCTC call of the past.  Is there value to identify this is the case, that to "apply probability and statistics to address uncertainty and risk relevant to civil engineering" is not just calculation/computation, or numbers?  Possibly this is something for the ASCE Commentary also.

    Early in my career, during a concrete pour, not a large quantity but for a foundation for a heavy tank, the one & only vibrator the contractor had on site stopped working.  Then I realized a bit more about why the specs. required a backup vibrator.  No computation/calculation or numbers were needed.

    Computation/calculation, numbers would seemingly always be a part of this criteria.  What stood out to me as a great example of construction engineering from many years ago was a retired construction company executive speaking about how many folks were convened to determine a safety factor.  The many folks included the company executive, other executives, lawyers, insurance company reps., & others.  The issue was NOT how to compute any value.  The issue was should a safety factor of 2.75 or 3.25 or something else be used because "failure could mean the end of the company."   Performing the computation/calculation is less an issue compared to the judgment of what safety factor is appropriate to use in the computation/calculation.

    Possibly such a distinction is also a matter of technology or engineering [ETAC or EAC].  At the moment, this discussion is EAC ABET relevant.  That makes me curious if apply is the appropriate Bloom's level.  An engineer should analyze and evaluate risk, in my opinion not just calculate/compute.  Ha, ha to me, when I search for Bloom's verbs, I get a site that has the word "calculating" first under the Analyzing column while "judging" is under the the Applying column.

    ------------------------------
    David Devine P.E., L.S., M.ASCE
    Fort Wayne IN
    ------------------------------



  • 5.  RE: _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted 06-03-2021 10:25 AM
    Dave,

    These are great insights.   Thanks for posting.

    In my view, we need to consider the two sections of this criteria provision separately.  I think that "apply probability and statistics" is inherently quantitative.  A probability is a number; a statistic is a number--so I don't think it's possible to apply probability or statistics non-quantitatively.  (Note that that these numbers might be estimated rather than calculated precisely--but they're still quantitative.)  But "addressing uncertainty and risk" is inherently judgment-based and thus is not necessarily quantitative.  Thus, I would certainly not interpret this provision as preventing qualitative risk assessments from satisfying the criterion.

    This is, indeed, an important issue for us to address in the Commentary.  Thanks again for raising it.

    ------------------------------
    Stephen Ressler Ph.D., P.E., Dist.M.ASCE
    Professor Emeritus
    Bethlehem PA
    ------------------------------



  • 6.  RE: _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted 06-03-2021 10:46 AM
    I think what truly matters is that, beyond the ability to compute numbers (i.e., repeat procedures), students can "see" the uncertainty in CE problems, and also that they understand the use of the procedures.

    It is not enough to be able to compute a mean and a median for a sample, which is what most students learn to do. They also need to understand in which cases it is a better idea to use the median instead of the mean; this is what most students are not able to do, even after a class in probability in statistics.

    ------------------------------
    Claudio Meier Ph.D., Ing., M.ASCE
    Associate Professor of Civil Engineering
    University of Memphis
    Memphis TN
    ------------------------------



  • 7.  RE: _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted 06-04-2021 08:02 AM
    I would follow on to Steve's comments on this.  As one who devotes a lot of time in risk analysis (much in areas beyond civil engineering), there are many sources of risk and uncertainty, e.g.:
    -- parametric (captured by standard prob/stat techniques)
    -- model (think of estimating stormwater flows using multiple different approaches)
    -- known unknowns (we don't have a "true" value of an input so we need to estimate it by sampling with finite resources)
    -- systematic changes (e.g. climate change)

    etc.

    two good references if anyone wants to go down this rabbit hole:
    Uncertainty: A Guide to Dealing with Uncertainty in Quantitative Risk and Policy Analysis

    Risk Modeling, Assessment, and Management (Wiley Series in Systems Engineering and Management)


    There is also the question of overall project risk, which is often addressed in the context of project management.

    ------------------------------
    Charles Haas F.ASCE
    LD Betz Professor of Environ. Eng. & Department Head - Civil, Architectural and Environmental Eng.
    Philadelphia PA
    [Phone]
    ------------------------------



  • 8.  RE: _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted 4 days ago

    I am not convinced that "address" is the intended word here.  Address implies mitigating uncertainty or risk.  Do we really expect students completing an undergraduate degree to be at that level?  If that is the expectation, rewording should be considered because probability and statistics are tools used to help a person develop plans to address uncertainty and risk but are not the tools used directly to address uncertainty and risk.  If the intent is actually to capture the CEBOK2, should the criteria be "…apply probability and statistics to determine [or quantify] uncertainty and risk relevant to civil engineering"?   



    ------------------------------
    James Hanson P.E., M.ASCE
    Associate Professor
    Rose-Hulman Institute Of Technology
    Terre Haute IN
    ------------------------------



  • 9.  RE: _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted 4 days ago
    Jim,

    Very insightful comment.  Thanks for your thoughtful contribution to this Forum.  The Task Committee will definitely consider your point as we move forward with an updated drat of the Program Criteria.

    Steve Ressler

    ------------------------------
    Stephen Ressler Ph.D., P.E., Dist.M.ASCE
    Professor Emeritus
    Bethlehem PA
    ------------------------------



  • 10.  RE: _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted 2 days ago

    I'm admittedly late in this conversation. With that said, I think the proposed change unnecessarily conflates and confuses important issues. I spent my career in the oil and gas industry where we routinely dealt with probability, statistics and risk. I think risk is the least understood of these concepts, but maybe the most important. Furthermore, a risk analysis does not require a calculation and can be done with a simple Risk Assessment Matrix, Link to example. Risk management does not require the application of probability and statistics Finally, and maybe most importantly, I would submit risk management is central to civil engineering and most everything we do as civil engineers. Risk management can involve probability and statistics, but it's not a requirement. 



    ------------------------------
    Mitch Winkler P.E., M.ASCE
    Houston, TX
    ------------------------------



  • 11.  RE: _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted 2 days ago
    With all due respect, as someone who spends the bulk of their time on risk assessment, statistics and probability are essential to do this at a high level.  Risk matrices, while often used, are rudimentary and conceal important aspects, and in fact (there is literature on this) may result in poorer decision making.

    ------------------------------
    Charles Haas F.ASCE
    LD Betz Professor of Environ. Eng. & Department Head - Civil, Architectural and Environmental Eng.
    Philadelphia PA
    [email protected]
    ------------------------------



  • 12.  RE: _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted yesterday
    Quantitative risk assessment has its time and place and my comment was not meant to demean or sideline its importance. However, I think there's an AND at play here. Ideally, engineers should be able to know the limits of a risk matrix AND know when more quantitative tools are needed. My point, to try and restate, is that engineers must have a fundamental understanding of risk and risk matrices are a simple and oft effective way of creating this understanding. This can be done with a basic understanding of statistics and probability in my opinion.

    ------------------------------
    Mitch Winkler P.E., M.ASCE
    Houston, TX
    ------------------------------



  • 13.  RE: _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted 2 days ago
    Thank you, Mr. WInkler, for your thoughtful comment about the nature of risk management and the limited relationship of risk management to probability and statistics.  The insight that you have provided is helpful and will be considered by the CEPC Task Committee.

    ------------------------------
    David Dzombak, PhD, PE, Dist.M.ASCE
    Hamerschlag University Professor and Dept Head
    Dept of Civil and Environmental Engineering
    Carnegie Mellon University
    Pittsburgh, PA 15213
    [email protected]
    ------------------------------



  • 14.  RE: _5. ... apply probability and statistics to address uncertainty and risk ...

    Posted 2 days ago
    While this is dealing with one part of Civil Engineering , Structural and Geotechnical engineering,  risk assessment is becoming more important.  For example on the statistic side in material,  understanding   how characteristic strengths are calculated and evaluated as well as how loads are set is fundamental  in being knowledgeable in Load Factored resistance Design and even more so in performance based design  promoted by the Structural Engineering Institute.   The demands for better designs in these areas are driving the need for understanding probability and statistics.

    ------------------------------
    David Thompson P.E., M.ASCE
    Principal
    KTA Structural Engineers Ltd.
    Calgary AB
    ------------------------------