Bojan Srdjevic

An Objective Multi-Criteria Evaluation of Water Management Scenarios

Advanced computer models are commonly used to simulate reservoir systems performance. If the number of possible management scenarios is large, it can be extremely difficult to follow related systems operation and get a valuable picture on its spatial and temporal behavior. The decision maker or analyst can be overburdened by quantity and complexity of information generated by model, particularly if system operation is repeatedly simulated for multiyear periods. Related problem is how to select the scenario with most desired long-term consequences. Possible approach is to use selected parts of models output and re-interpret system behavior by means of certain performance indicators, create appropriate decision matrix and perform multi-criteria analysis to rank decision alternatives (scenarios). The paper proposes a methodology that includes: (1) multiyear simulations of system operation; (2) computing spatially and temporally distributed systemperformance indices such as supply reliability, resiliency and vulnerability; (3) unbiased entropy-based weighting the importance ofperformance indices; and (4) final ranking of scenarios by means of multi-criteria analysis. The number of scenarios and number of performance indices is not restricted, and to account for possibly large sets of scenarios, an ideal-point-distance multi-criteria method TOPSIS is suggested. Proposed methodology appeared to be confident and robust in proof-of-concept application in Brazil.

Linking analytic hierarchy process and social choice methods to support group decision-making in water management

The social choice (SC) theory is in close relation with multicriteria decision-making (MCDM), especially in group decision contexts. SC theory includes various voting systems while MCDM is represented by utility and outranking methods; among utility models, the analytic hierarchy process (AHP) is probably the most popular in group decision support. In this paper, we investigate two possible contexts in modeling decentralized decision problems in water management. The first is based on AHP only and two group aggregation techniques. The second one assumes the AHP application in subgroups, while at a group level, aggregation is performed by the SC voting procedures. Comparative analyses show good agreement of the results when two methodologies are applied as the decision support to the water committee of the San Francisco river basin in Brazil. The second methodology (called AHP+SC) is considered more promising for implementation in real-decision situations in water management.

Identifying the Criteria Set for Multicriteria Decision Making Based on SWOT/PESTLE Analysis: A Case Study of Reconstructing A Water Intake Structure

This paper proposes an approach for defining the criteria set required for multicriteria decision making. An approach is developed for a specific class of water management problems, and a SWOT/PESTLE analysis is recommended for identifying the internal and external factors that influence a given water system. The factors are grouped into six categories: political, economic/financial, social, technical, legal, and environmental (PESTLE), and separated afterwards according to their positive or negative influence on the system. All factors are filtered by a proposed elimination algorithm to identify the non-inferior factors and declare them as candidates for inclusion into the criteria set. An approach is applied to the real-life problem of how to define the criteria set and enable the selection of the optimal reconstruction solution of a water intake structure within a regional hydro-system in Vojvodina Province, Serbia. To verify the developed approach, an independent expert is invited to asses all factors identified by the SWOT/PESTLE analysis using the analytic hierarchy process (AHP) and to rank factors by order of preference. The outcome is satisfactory because the seven top-ranked factors from AHP completely matched the list of factors derived from the elimination algorithm.

Approach of Decision Making Based on the Analytic Hierarchy Process for Urban Landscape Management

This paper proposes a two-stage group decision making approach to urban landscape management and planning supported by the analytic hierarchy process. The proposed approach combines an application of the consensus convergence model and the weighted geometric mean method. The application of the proposed approach is shown on a real urban landscape planning problem with a park-forest in Belgrade, Serbia. Decision makers were policy makers, i.e., representatives of several key national and municipal institutions, and experts coming from different scientific fields. As a result, the most suitable management plan from the set of plans is recognized. It includes both native vegetation renewal in degraded areas of park-forest and continued maintenance of its dominant tourism function. Decision makers included in this research consider the approach to be transparent and useful for addressing landscape management tasks. The central idea of this paper can be understood in a broader sense and easily applied to other decision making problems in various scientific fields.

Bi-criteria evolution strategy in estimating weights from the AHP ratio-scale matrices

Abstract The problem of deriving weights from ratio-scale matrices in an analytic hierarchy process (AHP) is addressed by researchers worldwide. There are various ways to solve the problem that are generally grouped into simple matrix and optimization methods. All methods have received criticism regarding the accuracy of derived weights, and different criteria are in use to compare the weights obtained from different methods. Because the set of Pareto non-dominated solutions (weights) is unknown and for inconsistent matrices is indefinite, a bi-criterion optimization approach is proposed for manipulating such matrices. The problem-specific evolution strategy algorithm (ESA) is implemented for a robust stochastic search over a feasible indefinite solution space. The fitness function is defined as a scalar vector function composed of the common error measure, i.e. the Euclidean distance and a minimum violation error that accounts for no violation of the rank ordering. The encoding scheme and other components of the search engine are adjusted to preserve the imposed constraints related to the required normalized values of the weights. The solutions generated by the proposed approach are compared with solutions obtained by five well-known prioritization techniques for three judgment matrices taken from the literature. In these and other test applications, the prioritization method that uses the entitled weights estimation by evolution strategy algorithm (WEESA) appears to be superior to other methods if only two, the most commonly used methods, are applied: the Euclidean distance and minimum violation exclusion criteria.

Robustness of AHP in selecting wastewater treatment method for the coloured metal industry: Serbian case study

Selection of the most suitable wastewater treatment method is a difficult task that requires the assessment of alternative methods across usually conflicting criteria. The analytic hierarchy process (AHP) is applied to evaluate four possible methods (chemical treatment, evaporation, separation by the use of the membranes, and biological treatment) applicable to the manufacture of coloured metals in Serbia. Seven typical criteria are used in the assessment of the treatment methods: energy consumption, price of the chemicals, effectiveness, simplicity of the process, price of the facilities, ecological impact, and necessary educational level of the workers. Following the standard AHP application to identify the most desirable treatment method, a sensitivity analysis is performed to explore the robustness of the final solution (the method to be applied in the factory) if ecological impact and energy consumption receive an increased relative weight compared to other criteria.

Heuristic aggregation of individual judgments in AHP group decision making using simulated annealing algorithm

SA algorithm is used as aggregation procedure in AHP group decision making.The group priority vector is obtained by minimization of the Euclidean distance.Objective search for maximum consensus between individuals within the group.Proposed procedure is invariant to any prioritization method.Proposed procedure performs better or at least equally to several other procedures. There are various aggregation procedures for obtaining a group priority vector within Analytic Hierarchy Process (AHP) supported decision making processes. This paper will introduce a heuristic aggregation procedure based on simulated annealing (SA) algorithm to be used for the purposes of obtaining a group priority vector at any node of an AHP hierarchy. The proposed procedure performs its aggregation process by minimizing the group Euclidean distance (GED) (consensus measure) across group weights and judgments, and the group vector obtained in this way is invariant to any prioritization method. In other words, there is no need to have individual priority vectors as is required by some other aggregation procedures. Along with SA minimization of the GED, the group rank reversal (minimum violation) criterion is implemented as a control mechanism, as well as the consensus measure based on the ranks of alternatives. The presented procedure is compared with several reported combinations of different prioritization methods and group aggregation procedures. Five examples from literature are used to show that the proposed procedure performs better or at least equally to several other well known combinations of prioritization and aggregation in AHP group decision making frameworks.

Virtually combining the analytical hierarchy process and voting methods in order to make group decisions

Multi-criteria decision making (MCDM) is represented by utility and outranking methods. Of the utility models, the analytic hierarchy process (AHP) is probably the most popular in group-decision support. The social choice theory (SCT) with its voting systems can be efficiently combined with MCDM, and AHP in particular, in various group-decision contexts. This paper investigates two possible contexts where modeling group decision-making processes in the field of human–computer interaction (HCI) takes place. Being an important part of universal access in the information society, the first context assumes the use of AHP only, associated with two different aggregating techniques to derive the group decision, with and without a consensus. The second context is based on using only SCT systems, i.e., preferential and non-preferential voting systems. The third, novel, approach that is proposed assumes a full AHP application in the first stage to obtain the weights of alternatives and, in the second stage, an interpretation of the AHP’s cardinal information as being the ordinal one and the direct application of the SCT voting systems. Comparative analyses show good agreement for the results when the three methodologies are applied as the decision support to ranking by importance (for a group of 14 PhD students) several widely used sources of information for the internet. The method of virtually combining the AHP and SCT voting systems could be efficiently implemented in real decision-making situations in HCI and related sectors, as well as in cross-sector settings.

Introducing the social participation concept in water management in Serbia, and related decision-making issues

This paper presents results obtained during recent studies of legal, financial, and organizational prerequisites to establish the water user associations (WUAs) in Vojvodina Province in Serbia. A procedure for launching a small-scale WUA is also proposed based on field data about two-fold irrigation and drainage systems. At a proof-of-concept level, the decision-making process is simulated by assuming the water users’ assembly as a group of representatives. By combining a multi-criteria method analytic hierarchy process with a social choice (voting) Borda method count, it was shown that a convincing decision-making framework is possible to establish and produce good final solutions (decisions).

An Extension of the Sustainability Index Definition in Water Resources Planning and Management

The sustainability index (SI) is a relatively new concept for measuring the performance of water resource systems over long time periods. Its definition is aimed at providing an indication of the integral behaviour of the system with regards to possible undesired consequences if misbalance of available and required waters occurs. SI is initially defined as a product and later reformulated as a geometric mean of performance indicators: reliability, resilience and vulnerability. As an extension of a recently published methodology to compute and use SI, in this paper we propose introducing two more indicators of system performance: (1) reliability of annual firm (safe) water as a system yield and (2) deviation of reservoir levels from corresponding rule curves. The last indicator is of particular importance if there are multi-purpose reservoirs in the system because reservoirs are the most important and sensitive regulators of the water regime within the system. We also propose a framework for assessing system performance in a systematic manner to compute SI at various locations within the system if different operating strategies are applied and, finally, how to evaluate strategies according to the resulting SI by using multi-criteria methods. A case study example from Serbia is used to illustrate the results of measuring sustainability under alternative operating scenarios for a system with three reservoirs and two diversion structures.