Optimal selection of fertilisers by horticultural consultants : a thesis presented in partial fulfilment of the requirements for the degree of Masters of Applied Science in Agricultural Systems and Management at Massey University

Abstract

Horticultural consultants recommending fertilisers must consider a large amount of information to determine an optimal fertiliser mix. Difficulty arises when matching the estimated levels of nutrient requirements with those available in fertiliser products, as it is unlikely a single, or combination, of fertilisers will match exactly the required nutrients. It is assumed that consultants are proficient at making fertiliser recommendations, but desire to improve the process of making, and accuracy of, the recommendations. The objectives of this study were to describe the process consultants go through when providing a fertiliser recommendation and then use this information to design a decision support system (DSS) as an aid to the fertiliser selection process. Data was collected from three consultants to develop a conceptual model of how the fertiliser recommendation service consultants provide operates. With this information a DSS model consisting of client and fertiliser databases, an optimisation component and an user interface was developed. The optimisation component uses a compromise programming approach. The objective function minimises the deviations from the targets of a cost goal and nutrient requirement goals. The DSS was run using samples provided by the consultants and the results were compared with what they had recommended. Even though the results from the DSS model did not match exactly what the consultant had recommended, most of the DSS solutions were cheaper and weighed less. The information generated provides a good starting point from which to make a fertiliser mix recommendation by combining other factors such as pH and solubility of each fertiliser, and other rules of thumb that consultants use.