Integration, participation and optimal control in water resources planning and management

https://doi.org/10.1016/j.amc.2007.09.069Get rights and content

Abstract

Water resource systems provide a challenging testing ground for System Analysis and Optimal Control Theory. The intrinsic complexity of the physical system, which is highly non-linear and affected by strong uncertainty, intermingles with the value-laden socio-economic context in which decisions on the systems have to be taken. Mathematical models and algorithms have to be re-considered within a methodological framework, in which stakeholder participation and cross-disciplinary approaches are given a central role. In this paper the Integrated and Participatory Planning (PIP) procedure developed by the authors is presented as a potential methodological approach to the effective and efficient planning and management of water systems. Great emphasis is given to the way to combine traditional control techniques with preference and subjective aspects of decision-making. The key role of Multi-Objective Decision Support Systems (MODSS) is also stressed as a way of bridging the gap between theory and practice.

Introduction

Integrated Water Resource Management (IWRM, [15]) is emerging as an accepted alternative to the sectoral, top-down management style that has disastrously dominated in the past. IWRM is based on the recognition that the intrinsic complexity of interconnected biophysical, social, economic and political factors can only be addressed by combining and truly integrating social constructivist ideas of participation and empowerment with a cross-disciplinary engineering approach. Water conflicts are value-laden and not neutral in a technical sense, and thus they cannot be resolved in a fully rational decisional context: the combined presence of multiple water uses and strong vested interests in the water resource management, both within and between nation states, requires the active involvement of the stakeholders in each stage of the decision-making process, from the identification of their preference structure to the negotiations of the decision(s) to be actually implemented. On the other hand, the existence of many alternative decisions and the uncertainty on their future effects require a systematic and formal approach through which the most interesting (namely Pareto efficient) decisions are singled out and their effects somehow assessed ex ante.

To couple effectively technical issues and preference aspects a procedural guidance must be provided to the decision-making process and an appropriate (ICT) toolbox designed to support planning as a systematic, integrative and iterative process [12]. According to these requirements, a general procedure for Participatory and Integrated Planning (PIP) has been conceived and implemented by the authors [9] by formalizing the methodologies developed and experienced in the Verbano Project (EU-INTERREG II), and further tested, completed and validated in a number of other international projects (MERIT, HARMONI-CA and TWOLE). A Multi-Objective, web-based, Decision Support System (TWOLE) has been designed to support each phase of the procedure.

The purpose of this paper is to review the PIP procedure in a strict System and Optimal Control Theory perspective, showing in particular how traditional and innovative control techniques play a central role in its application and how they can be re-interpreted to cope with the preference and subjective aspects of the decision-making process. The need for effective Multi-Objective Decision Support Systems (MODSSs) is also stressed and their key features analysed. Finally future and emerging research topics are presented.

Section snippets

The PIP procedure

Our Participatory and Integrated Planning (PIP) procedure provides a conceptual framework for the development of the decision-making process in the presence of multiple objectives and multiple Decision Makers (in the following DMs). A detailed description of the PIP procedure in its theoretical and applicative aspects can be found in [28], [29], respectively. The PIP procedure has been conceived to be as flexible and general as possible, and thus it allows for the use of both quantitative and

The Multi-Objective Decision Support System

The implementation of each phase of the PIP procedure requires the use of specific ICT tools. If we go through the PIP procedure, we can identify a number of tasks that can be helpfully supported by dedicated software programmes. For example, the definition and validation of the indicators (phase 2) is easier if the stakeholders are given the possibility of quickly computing and comparing the indicator values over historical data sets and perform sensitivity analysis. The identification of the

Policy design

Each phase of the PIP procedure presented in Section 2 poses a number of theoretical and practical questions; however the interest will here be focused on phase 4 only, to be precise on the solution of the design Problem P1.

Conclusion and future directions

The paper presents a procedure, the PIP procedure, which lies at the crossroads of different disciplines and aims at integrating stakeholder participation with Optimization and Decision Theory methods in water resources planning and management. In fact, if on the one hand the active involvement of the stakeholders is a necessary condition for effective and equitable planning and management of environmental systems, on the other hand the application of methods from Optimization and Decision

Acknowledgement

Partially supported by FONDAZIONE CARIPLO TWOLE-2004.

References (38)

  • A. Castelletti, G. Corani, A.E. Rizzoli, R. Soncini-Sessa, E. Weber, A reinforcement learning approach for the...
  • V.M. Johnson et al.

    Accuracy of neural network approximators in simulation–optimization

    Journal of Water Resources Planning and Management

    (2000)
  • European Commission, Common Implementation Strategy for the Water Framework Directive, Guidance Document no. 11:...
  • A.P. Georgakakos

    Decision support systems for integrated water resources management with an application to the Nile Basin

  • K.C. Gilbert et al.

    TVA hydroscheduling model: theoretical aspects

    Journal of Water Resources Planning and Management-ASCE

    (1982)
  • Global Water Partnership (GWP), Integrated water resources management, TAC Background Paper 4, GWP Secretariat,...
  • E.R. Hooper et al.

    Optimal stochastic operation of Salt River Project, Arizona

    Journal of Water Resources Planning and Management-ASCE

    (1991)
  • L.P. Kaelbling et al.

    Reinforcement learning: a survey

    Journal of Artificial Intelligence Research

    (1996)
  • R.L. Keeney et al.

    Decision with Multiple Objectives: Preferences and Value Trade-offs

    (1976)
  • Cited by (287)

    • Integrated optimal allocation model for complex adaptive system of water resources management (I): Methodologies

      2015, Journal of Hydrology
      Citation Excerpt :

      Srdjevic et al. (2004) used an integrated simulation operation, system performance indices and multi-objective evaluation model to simulate operation of reservoir system. Castelletti et al. (2008) used multi-objective decision to cope with the preference and subjective aspects of the water resources allocation. Shourian et al. (2008) used a multi-objective model to allocate water resources optimally over time and space among competing demands based on particle swarm optimization algorithm.

    View all citing articles on Scopus
    View full text