Fuzzy multiple-criteria decision making for crop area planning in Narmada river basin

The present study dealt with the real-world problem of irrigation water management of evolving suitable cropping pattern, which should be in harmony with optimal operation of the multi-reservoir system in the basin. The Narmada river basin system, comprising 11 reservoirs either existing, planned or under construction, has been first simulated and analyzed. The time horizon used to simulate the monthly operation of the system corresponds to the past 30 years of history. A multi-objective fuzzy linear programming (MOFLP) area allocation model has been formulated to cope with the diverse/conflicting interests of diAerent decision makers such as the irrigation authority (government) and the individual farmers involved. Simulation output in the form of optimal monthly releases for irrigation is one of the main inputs to the MOFLP area allocation model. Variable irrigation demand over the planning time horizon has been incorporated into the formulated model considering high variation in precipitation. Thus, varying cropping patterns in the command area, one for each year, have been analyzed which are based on the past 30 years of reservoir simulation. Besides this, a cropping pattern corresponding to 80% dependable releases and rainfall is also analyzed. Such analysis is extremely useful in deciding on an appropriate cropping pattern in any command area that minimizes the average crop failure risk in view of uncertain irrigation water availability, especially in dry years. # 2000 Published by Elsevier Science Ltd. All rights reserved.