Education Needs and Technology Transfer
This chapter addresses Question 4 as defined in Chapter 1, namely: How should probabilistic methods be taught in civil engineering curricula in universities? Are the requirements of geotechnical engineering so unique that special considerations are required for this area? How should probabilistic methods be introduced to practicing geotechnical engineers who have had little or no formal education in the underlying theory?
Based on the input received during the workshop, the committee has identified a number of changes that would contribute to the education of civil engineers in probabilistic methods. Among the most important is the creation of mechanisms and incentives for both educators and practitioners to learn these new approaches to problem solving. Probability concepts, particularly as they pertain to reliability and risk assessment, should be included in statistics courses that are now required in U.S undergraduate civil engineering curricula. Further, probabilistic consideration should be emphasized in continuing education courses and other professional forums. These actions would help to provide the essential background for practitioners to judge the types of problems for which probabilistic methods can yield an appropriate solution and those for which they cannot, as well as providing practitioners with the tools to implement the methods when appropriate.
EDUCATION OF GEOTECHNICAL ENGINEERING STUDENTS
Probabilistic methods, including applications to reliability and risk assessment, should be introduced into the civil engineering curriculum at the earliest possible time. In most curricula, freshman or sophomore courses in statistics are already in place, but emphasis on probability and the application of probability theory to risk assessment exists only at a few educational institutions. The establishment of such courses falls within the current guidelines for civil engineering of the Accreditation Board for Engineering Technology. At most institutions, integrating probability into the curriculum will require only the reorientation of existing courses and will not require additional course credits in the curriculum. The cooperation and leadership of the departmental governance will be
needed, however, because such courses will necessarily be available to all civil engineering students, not just those planning to specialize in geotechnical engineering.
Simultaneously, a systematic and comprehensive effort must be made either by department heads or by geotechnical group leaders to assure that appropriate follow-up instruction is initiated. For example, application of probabilistic methods to the reliability evaluation of simple, well-defined problems can be taught in elementary courses in geotechnical engineering, and an introduction to more-complex problems, such as waste remediation or rock-fall management, can be provided in comprehensive design experience courses, most appropriately in those that integrate geotechnical engineering with other disciplines.
Most leading civil engineering departments already have at least one course on advanced applications of probabilistic methods, covering topics such as reliability, stochastic processes, and decision analyses. These courses have emphasized structural and hydraulic engineering applications. Probability concepts and techniques that are unique to some of the geotechnical problems, such as probabilistic site characterization and the use of Bayesian decision theory to model the observational method, are not discussed. It is therefore desirable to have a similar advanced probabilistic methods course focusing on geotechnical applications. Case histories of actual projects that involve the use of probabilistic methods should be included, either in this course or, better still, in conventional geotechnical courses, to provide the perspectives of actual project experiences.
The implementation of these recommendations will be difficult because of the scarcity of faculty with appropriate background in probability theory and experience in applications, coupled with a reluctance on the part of faculty to develop new course materials that carry little foreseeable reward per se. In this regard, as probabilistic methods are increasingly used in geotechnical practice, instructors will appreciate the value of applying these methods, and they will be more motivated to teach them. The process can be accelerated by increased funding of university research in geotechnical reliability and of applications by government agencies and by industry.
Another impediment to the implementation of the recommendations is the lack of textbooks, software, and other instructional materials appropriate for both undergraduate and graduate students that have specific applications to the numerous facets of geotechnical engineering. Several texts (e.g., Benjamin and Cornell, 1970; Ang and Tang, 1975, 1984; Harr, 1977, 1987) are available that deal with probabilistic and decision methods in the general area of civil engineering. Although the basic concepts and methodologies are well presented and demonstrated with simple engineering examples in these texts, more examples of in-depth geotechnical applications are needed. These applications should include, for example, detailed case studies and probabilistic models that are specifically useful for the modeling of soil properties.
If appropriate academic courses are to be instituted, teaching materials should be available to professors on a nationwide basis to facilitate the course development. A mechanism for rapid information transfer to geotechnical engineering educators would be short courses targeted specifically for that audience. Such short courses would feature faculty experts from leading universities who would lecture and lead discussions on
probability theory and applications and provide master copies of materials such as handouts, software, case histories, problems and solutions, and video and CD-ROM resources.
Workshops for training geotechnical faculty on probability concepts and methods are also important. The National Science Foundation is a likely sponsor of these workshops, and sponsorship by other government agencies and professional groups also could be explored. Qualified instructors for such a faculty workshop on probabilistic methods in geotechnical engineering could be drawn from the participants in the 1992 reliability workshop. New textbooks and educational software are likely outputs of such proposed faculty workshops.
EDUCATION OF PRACTICING GEOTECHNICAL ENGINEERS
The traditional approach taken by universities in the continuing education of practicing geotechnical engineers may not be adequate to achieve the fullest assimilation of probabilistic methods into geotechnical practice. Practicing geotechnical engineers who have not had the opportunity for a formal education in probabilistic methods will need to be motivated and given the tools to apply probability theory. The benefits of appropriate application of probabilistic methods will need to be convincingly demonstrated to practicing geotechnical engineers. The education of practitioners and managers is an important component in the conversion of probability theory from a subject of narrow interest to a topic that is an integral part of geotechnical engineering practice and that is used like the finite-element method or any other family of analytical tools when appropriate situations arise.
Universities that develop strong programs in probabilistic methods for geotechnical engineering should provide continuing education for practicing geotechnical engineers. This should be done in cooperation with those practicing geotechnical engineers proficient in probabilistic methods. Short courses for practicing geotechnical engineers will provide needed discretionary funds for the universities, which could be a strong motivating factor to develop and implement such courses. Short courses can be of the traditional in residence kind or may be offered cooperatively with government agencies or industry through satellite or remote-site TV cable systems. Short courses on probabilistic methods for geotechnical practitioners and researchers are, in fact, currently available within specific organizations (e.g., the U.S. Army Corps of Engineers and Sandia National Laboratories). Such courses might be modified for the purpose indicated here. Professional organizations such as the American Society of Civil Engineers or the American Society of Foundation Engineers should be encouraged to sponsor and organize continuing-education activities in probabilistic methods and risk mitigation in geotechnical practice. As continuing-education courses on probabilistic methods for geotechnical engineering are developed and refined, they can be videotaped, and the tapes distributed to small engineering firms and to universities with smaller geotechnical engineering programs.
For a rapid advancement of the use of probabilistic methods by practitioners, it may be desirable to the geotechnical profession to have acknowledged state-of-the-practice
probabilistic methodologies for frequently encountered geotechnical problems. In this respect, technical committees (for example, committees of the American Society of Civil Engineers) should be encouraged to organize sessions at professional conferences that specifically present state-of-the-practice procedures. This could be accompanied by the publication of design manuals containing step-by-step analysis and design procedures. Computer programs that are widely used to assess the required calculations could be made easily available. However, such programs should be well-documented in terms of the underlying theory, assumptions, and methodology; the users should be aware of any limitations that the programs may have.
To accelerate the learning process of the practitioners, the committee suggests that, to the extent possible, major geotechnical projects should involve a probability expert as part of the project team to provide opportunities for close interaction between that expert and the other team members. For example, the probability expert could assist team members with analysis and design procedures. With the increasing involvement of probability experts in various geotechnical projects, publication of those case studies will serve to demonstrate examples of probabilistic applications that are acceptable to, and usable by the profession. Once the owners of the projects are made aware of the potential benefits of probabilistic approaches through these examples, they would begin to require the use of such approaches in geotechnical projects. This, in turn, will provide further reason for implementation of probabilistic methods. A good example of the use and benefits of probability theory that follows from these considerations can be seen its application in offshore geotechnical engineering (see Wu et al., 1989; Tang and Gilbert, 1993). A specific example of this application was presented as Example 8 in Chapter 2.
Ang, A. H-S., and W.H. Tang. 1975. Probability Concepts in Engineering Planning and Design, Vol. 1: Basic Principles. New York: John Wiley & Sons.
Ang, A. H-S., and W.H. Tang. 1984. Probability Concepts in Engineering Planning and Design, Vol. II: Decision, Risk and Reliability. New York: John Wiley & Sons. Currently distributed by BookMasters, Inc., Ohio.
Benjamin, J.R., and C.A. Cornell. 1970. Probability, Statistics and Decision for Civil Engineers. New York: McGraw-Hill.
Harr, M.E. 1977 Mechanics of Particulate Media. New York: McGraw-Hill.
Harr, M.E. 1987. Reliability-Based Design in Civil Engineering. New York: McGraw-Hill.
Tang, W.H. and R.B. Gilbert>. 1993. Offshore Pile System Reliability. Pp. 677–686 in Proceedings of Offshore Technology Conference, OTC 7196. Texas: Offshore Technology Conference.
Wu, T.H., W.H. Tang, D.A. Sangrey, and G.B. Baecher. 1989. Reliability of offshore foundations—State-of-the-art. Journal of Geotechnical Engineering, American Society of Civil Engineers 115(2): 157–178.