Colin B. Brown

Professional decisions: the central role of models

Engineering is largely concerned with models and modelling. Models provide a context in which decisions are made. Here, the modelling process is considered in a general sense and then the relationship with engineering models is developed. For understanding, a first step is to differentiate and categorise. Thus, 10 types of engineering models are proposed and their purposes with respect to professional decision-making are discussed. The role of models in engineering failures is also considered.

Professional decisions: responsibilities

The responsibilities which civil engineers, and particularly the engineer of record, accept are considered. The interaction with other decision-makers such as the owner, stakeholders, the law and contractors results in complexity that can be partially resolved by the introduction of protocols in the form of regulations and codes of practice. However, uncertainty always exists and can result in surprises that can produce both beneficial and bad results. The sections entitled the cast, protocols and reality, advocacy and surprise, and complexity cover these topics. The nature of responsibility is analysed. Professional engineers work within an increasingly complex environment and have a responsibility to acquire and use skills beyond those applicable to traditional technical issues.

A general lower and upper bound theorem of static stability

Abstract A statically stable state of a system subjected to conservative and dissipative forces is considered as a local minimum of the sum of the potential energy and the energy dissipated from the system subject to the kinematic constraints on the system. This stability criterion is investigated by the methods of optimization under constraints. A dual mathematical program, the maximization of the complementary energy of the system subject to equilibrium contraints, is constructed. Bounds on the kinematic state space of a system and energy dissipation are introduced as inequality constraints. Lower and upper bound conditions for the loads causing instability of the system are derived. By the upper bound condition, the system is unstable if the virtual work is negative in a kinematically admissible displacement, including rigid body components. By the lower bound condition, the system is stable if the gradient vectors of the active constraints with nonzero Lagrange multipliers span the space of feasible rigid body rotations. The existence of a nonempty feasible set for the dual program is also found to ensure the stability of the system.

Intuitive decisions and heuristics – an alternative rationality

Two ways of making a decision are objectively using formal analysis and subjectively using intuition. Psychological research has shown that the latter leads to better decisions in complex situations. Intuitive decision-making uses simplified rules of thumb heuristics. In engineering terms, complexity implies complex systems. A set of rules and principles – heuristics – is presented as a guide for making subjective decisions in complex system contexts, one aspect of which is to provide quality control and guard against bias and error. An extended practical example of the use of these heuristics is given involving assessment of the safety of nuclear-powered ships.

Engineering decisions: framework, process and concerns

Decisions are central to engineering processes and hold them together. It is argued that better decisions will lead to better engineering. To achieve better decisions requires that they be understood in detail. A typical decision is broken down into its essential requirements and processes, thus displaying the components of its framework. The process leads to the identification of a number of concerns. The components are discussed and a set of issues where more work needs to be done is identified. There are significant implications for both engineering practice and engineering education.

Optimizing and satisficing

Abstract Optimization has the appeal of mathematical elegance and the possiblity of being of practical significance. This significance is examined for the problem of producing least-cost structures. For aircraft, this simplifies to the production of structures of least weight, whereas in civil engineering, the existing conditions for design and construction in the United States seldom produce an environment for the existence of least-cost structures. Optimization procedures are examined within these situations when the objective functions and constraints are deterministic and when they are uncertain. The uncertainty is expressed as amenable to both probabilistic and fuzzy set specifications. This transformation, from crisp optimization to the concept of satisficing of structures, is explored.

Reliability in the context of design

Abstract The major purpose of this paper is to formulate the structural reliability problem in the context and objectives of the structural designer. In this formulation, the reliability of the structural system is represented in network form, allowing the designer to control global behavior of the system as a whole through the selection of a critical pathway through this network. This approach differs from that of a member-by-member local optimization scheme in that the selected pathway, representing a particular failure mode, is maximized, within certain bounds, relative to other modes. This type of approach to designing for global behavior is fundamentally old; yet, the reliability computations presented herein are new. Recent developments in object-oriented programming languages provide opportunities for effective computational modeling of these networks.

Functional Reliability of Structures

Abstract A general theoretical foundation of structural system reliability analysis is presented. Both an analytical and a numerical formulation for evaluating a relative reliability measure of a structural system is developed in which general random factors affecting both strength and loading can be included. The reliability of a suspension bridge is analyzed as an example; the effect of dead load, used as a proof load, is also studied.

Uncertainty in civil engineering systems: probability and fuzzy sets

Abstract Decisions in civil engineering systems have to be made in the face of uncertainty. Probability theory has been used to quantify this uncertainty. Claims for the use of fuzzy set theory in accommodating semantic ideas have been made. The relationship between these two theories as expressions of uncertainty in decision making is explored by means of Saatys two by two construction for the assessment of subjective probabilities and fuzzy supports, and by an examination of the psychological evidence with respect to fuzzy set theories.

Consistent crudeness principles in design and forensic decision making

Abstract Formal decision making involves the generation of alternatives that meet the constraints and objectives, followed by a comparison between the alternatives with respect to decision criteria. The process is systematic and hierarchical, proceeding from broad decisions to ones of great specificity. At each level in the hierarchy comparisons among alternatives are made, and a question arises concerning whether the analytical and information efforts are much the same for each of the competing alternatives. This problem is the concern of this paper where the evenhandedness is dealt with by principles of crudeness. Four principles are developed, and the application of these to professional decision problems in design and forensics are considered. The measures of crudeness are developed as quality statements about the variables, information and modeling of the-alternatives. The mathematical aspects are in terms of fuzzy sets theory and fuzzy entropy.