CommonKADS Model Framework for Web Based Agricultural Decision Support System

Authors

  • Jignesh Patel Nirma University, Ahmedabad, India, Institute of Technology, Nirma University,
  • Chetan Bhatt Government Engineering College, Rajkot, India

Abstract

Increased demand of farm productions and depleting natural resources compelled the agriculturecommunity to enhance the use of Information and Communication Technology (ICT) in variousfarming processes. Agricultural Decision Support System (DSS) proved useful in this regard asthe agricultural systems are complex and partially known. The majority of available AgriculturalDSSs are either crop or task specific. There are very less endeavors found in the direction ofcomprehensive DSS. The specific DSSs mainly developed with rule based or knowledge transferbased approach. These methodologies lack the ability to scale up and to support the developmentof large DSS. Modeling approaches are more suitable than so called transfer approaches for largeand inclusive DSS. Unfortunately, it is found that that the model based knowledge engineeringapproach is not much utilized for the development of Agricultural DSS. The modeling approachto construct Knowledge Base Systems (KBS) becomes well accepted among the KnowledgeEngineering (KE) communities due to its modular structure and ability to break down theknowledge engineering problem into smaller tasks. Modeling approach for the development ofDSS offers the broad idea of structure and modules of the support system before hand. There aremany modeling frameworks proposed and subsequently used by the KE communities.CommonKADS is one of the popular modeling frameworks for KBS. The paper presents theorganization, agent, task, communication, knowledge and design models based onCommonKADS approach for development of scalable, broad and practically usable agriculturalDSS. A web based DSS developed with multi agent CommonKADS modeling approach. Thesystem offers decision support for irrigation scheduling and weather based disease forecastingfor the popular crops of India. The proposed framework along with the required expertknowledge, provide necessary platform on which the larger DSS can built for any crop of givenlocations.

Downloads

Issue

Section

World Conference on Computers in Agriculture, San Jose, Costa Rica, 2014