Pierre Louis Kunsch

Strange: An interactive method for multi-objective linear programming under uncertainty

Abstract In the field of investment planning within a time horizon, problems typically involve multiple objectives, and basic data are uncertain. In a large number of cases, these decision problems can be written as linear programming problems in which time dependent uncertainties affect the coefficients and the right hand side of constraints. Given the possibility of defining plausible scenarios on basic data, discrete sets of such coefficients are given, each with its subjective probability of occurrence. The corresponding structure is then characteristic for Multi-Objective Stochastic Linear Programming (MOSLP). In the paper, an interactive procedure is described to obtain a best compromise for such a MOSLP problem. This algorithm, called Strange , extends the Stem method to take the random aspects into account. It involves in particular, the concepts of stochastic programming with recourse. In its interactive steps, the efficiency projection techniques are used to provide the decision-maker with detailed graphical information on efficient solution families. As an illustration of the successive steps, a didactic example is solved in some detail, and the results of a case study in energy planning are given.

Nuclear waste management: an application of the multicriteria PROMETHEE methods

Radioactive waste management is an important issue for every electronuclear programme. A particular problem is to know how to finance the waste disposal: indeed, the time span between the electricity production itself and the final disposal in geological formations can represent decades. The problem is made more complex by the choice of a time scenario and of a disposal site. Moreover, the decision-making process involves several points of view of different actors: electricity companies, consumers, public bodies, etc. A multicriteria analysis has therefore been applied, using the PROMETHEE methods and the geometrical representation GAIA. These methods are well adapted to this problem with many actions and few very conflicting criteria.

Combining multicriteria decision aid and system dynamics for the control of socio-economic processes. An iterative real-time procedure

The paper presents the elements of a new methodology to control complex and hypercomplex socio-economic structures. The control process is iterative, combining the principles of System Dynamics, Control theory and the PROMETHEE Multicriteria Decision Aid (MCDA) methodology. It consists of three main stages: setting up and calibration of a quantitative model, definition of long-term strategies and short-term control. The purpose is to define within a panel of decision makers appropriate strategies towards long-term goals, and to implement suitable control measures. These should in particular help cope with progressive and catastrophic variations in the behaviour of the system.

Nuclear fuel cycle optimization using multi-objective stochastic linear programming

Abstract The share of electricity produced by nuclear power plants is still increasing in many countries. If the head-end of the nuclear fuel cycle, regarding all operations before leaving the reactor, has now reached maturity, many pathways are still possible in the back-end of the cycle, dealing with the spent fuel. The long-term consequences of any decision regarding subjects such as fuel reprocessing, waste disposal, etc., require the use of strategic planning methods, in which provision is made for multiple objectives and for uncertainties. An interactive multicriteria approach using stochastic linear programming is provided by the code Strange , which has already been applied to different energy strategy problems, also outside the nuclear field. The application presented in the paper illustrates the search for a best fuel cycle policy including four criteria: production costs, the supply of raw material, the commercial balance and employment. The concept of scenarios is used to describe future uncertainties.

The importance of systems thinking in ethical and sustainable decision-making

Ethical values beyond the satisfaction of basic human needs are specific to each society at a given time. Modern societies are confronted to the challenges of disappearing natural resources, fierce competition on global markets, and climate change. In this paper we define ‘good’, and at the same time ‘ethical‘ decisions in the 21st century as being in accordance with the principles of Sustainable Development. Operations Research can assist sustainable decision-making in two ways: (1) through the tools of systems thinking, in particular system dynamics and soft system modelling, to provide insight into the way complex non-linear living systems and human societies function, and help making sustainable decisions, and (2) through educating and training young people in systems thinking. The paper presents examples of simple models that could serve in classrooms.

The zero-emission certificates: A novel CO2-pollution reduction instrument applied to the electricity market

Abstract Today several instruments exist to decrease air pollution. The paper discusses pollution taxes, emission trading permits and green certificates applied to reduce the CO 2 -emissions in the electricity sector. It then investigates how the different mechanisms behind these three instruments can possibly be combined. The proposed approach is to introduce the concept of zero-emission certificates (ZEC). ZEC are confirming actual emission reductions achieved by electricity producers as compared to a well-defined baseline. Producers can trade ZEC on a market to achieve least-cost efficiency in their reduction efforts. Distributors can themselves produce an additional contribution to emission-reductions by decreasing the final demand, i.e., by producing zero-emission MWh. In this way the electricity market is approached from both the supply and the demand sides. The paper uses system dynamics to validate the approach. It shows why it is in the interest of all operators to make the largest possible reduction efforts as long as they are compatible with economic efficiency.

Multi-Objective Decision Making under Uncertainty: An Example for Power System

In order to further rural development in the Third World, dispersed electrical power systems can be advantageously used. Small communities far removed from the grid expect to be able to produce electricity in the most reliable and self sufficient way with their own limited qualified manpower and primary ressources. Solar technologies systems may therefore be considered whenever high insolation is available, instead of using Diesel engines and expensive fuel. Moreover, the modularity and long lifetime of such systems are very desirable features. A traditional cost/benefit analysis based on production costs and investment risks is not yet favourable due to the high initial cost of solar systems. Nevertheless, if other criteria such as security of supply, balance of payments, etc ... are taken into account, solar energy can possibly be an interesting proposition.

Application of the ORESTE method to a nuclear waste management problem

Abstract Many multicriteria analysis methods require some detailed information concerning the different criteria: weights; order relation; preference functions; etc. However in many real cases, it is quite difficult to provide this information. The following paper presents a decision problem on nuclear waste managemnt, in while little information is available on the multiple criteria, representing each the point of view of a different actor: electricity producer, consumer, public body. For this type of decision problems, an outranking method like ORESTE seems particulaly well adapted because its only request in a preorder of the criteria. It has been applied and a sensitivity analysis is performed in order to provide that the results are largely independent of the assumed order relation on the criteria.

A fuzzy decision support system for the economic calculus in radioactive waste management

A methodology has been elaborated to derive contingency factors for evaluating in a realistic way the costs of first-of-the-kind projects in radioactive waste management (RWM). The paper describes the practical implementation of the fuzzy decision support system (FDSS) and its interface to assist economists in charge of the economic calculus. Uncertainties to be added on top of basic cost evaluations are represented by two contingency factors, respectively, called the P- and the T-factors. The P-factor represents the uncertainties of the project induced by its still incomplete advancement; the T-factor represents the uncertainties caused by the still insufficient technological maturity on which the project is based. Progressive implication rules of the Goguen type are used with the two contingency factors as outputs. Input variables for P and T are given on relative [0,1] scales. Fuzzy logic is also used as front-end for obtaining the two inputs in the course of peer reviews of technology-experts and project-specialists. To that aim, the semantic opinions and past experience of the latter are expressed in the form of conditional and unconditional rules. The credibility of T-experts and P-specialists are taken into account by using the Kleene-Dienes inference. A numerical example on the cost of disposal of high-level radioactive waste in a deep repository is used to illustrate the practical use of the FDSS. This approach is also applicable to other economic assessments inside or outside the nuclear field.

APPLICATION OF MULTI-OBJECTIVE STOCHASTIC LINEAR PROGRAMMING TO POWER SYSTEMS PLANNING

Abstract In order to further rural development in the Third World, dispersed electrical power systems can be advantageously used. Small communities far removed from the grid expect to be able to produce electricity in the most reliable and self sufficient way with their own limited qualified manpower and primary resources. Solar technologies using thermodynamic or photovoltaic systems may therefore be considered whenever high insolation is available, instead of using diesel engines and expensive fuel. Moreover, the modularity and long lifetime of such systems are very desirable features and it should prove possible to manufacture most components in the developing country itself, after a transition phase. A traditional cost/benefit analysis based on production costs and investment risks is not yet favourable due to the high initial cost of solar systems. Nevertheless, if other criteria such as security of supply, balance of payments, etc., are taken into account, solar energy can possibly be an interesting proposition. An interactive multi-objective analysis can then be developed to consider any number of criteria and to take into account the large future uncertainties associated with projected data available in the energy field. In the first part of this paper, a particular case is described in the context of small rural communities with important solar resources. In section II, the problem is completely modelled as a linear program with three objectives. Scenarios are considered on the basis of average and extreme assumptions of diesel fuel price increases and on solar technology cost reductions. An interactive procedure is applied in section III to resolve this multi-objective stochastic linear program and to obtain a satisfying compromise for the decision maker. A Fortran code, STRANGE, has been written and implemented: the results are presented in terms of the relative production of diesel or solar generators.