Evaluation of alternative sugarcane selection strategies
Calija-Zoppolato, Vanja (2004) Evaluation of alternative sugarcane selection strategies. PhD thesis, James Cook University.
|PDF (Thesis front) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader|
|PDF (Thesis - front pages; chapters 1-3) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader|
|PDF (Thesis - chapters 4-6; references; appendices) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader|
The international competitiveness and success of the Australian sugar industry, which is one of the world's largest exporters of raw sugar depends on increased cane yield and advanced farming practices. One of the key drivers for a sustainable sugar industry is therefore, to increase cane yield through designing efficient breeding programs, that aim at producing new and improved varieties of cane. Selection for superior genotypes is the most important aspect of sugarcane breeding programs, and is a long and expensive process. It consists of a number of stages where at each stage some genotypes are chosen for further selection and some are discarded from future selection. Designing a selection system is a complex task, with varying parameters at each stage. While studies have investigated different components of selection independently, there has not been a whole system approach to improve the process of selection.
The aim of this research was to develop a tool for the optimisation of selection systems. The problem of designing an efficient selection system has two components: firstly, evaluating the performance of selection systems and secondly, deciding on a combination of selection variables that will select the most promising genotypes. These two components were designated sub-objectives, one and two respectively.
To address the first sub-objective, data on previous selection trials was collected and used to predict gain for different selection designs. The value that is used to compare the performances of different selection systems is what was called in this thesis, the genetic gain for economic value (G), a measure based on the estimate of a potential economic value of a genotype if planted as a cultivar. The connection between (G) and choice taken for selection variables at various stages is complex and not expressible by a simple set of formulas. Instead, a computer based stochastic simulation model SSSM (Sugarcane Selection Simulation Model) was developed.
To eliminate as many simplifying assumptions as possible and bring the study as close to real life as possible, the quantitative genetics of sugarcanes relevant to selection was studied. Furthermore, a specific sugarcane-breeding region was targeted, the Burdekin region (Australia). To ensure the accuracy of the SSSM, it performance was verified and the sensitivity analysis was performed to identify those variance parameters to which it is most sensitive.
By, developing the SSSM this study approached the problem as an integrated system, where if one parameter changes the state of the whole system changes. Furthermore, by creating an accurate selection simulation model a new methodology for evaluating alterative sugarcane selection strategies was obtained. A new methodology that tests the performance of different selection designs prior to their field trials and also tests the impact any change in the estimated variance components may have on selection, will be a potential money saver for the industry. Furthermore, the SSSM can be directly applied to any region targeted by sugarcane breeding programs or to other clonally propagated crops.
The second sub-objective was addressed by the development of the optimisation algorithm called ASSSO (Algorithm for the Sugarcane Selection Simulation Optimisation), a combination of dynamic programming and branch-and-bound. The ASSSO was applied to the Burdekin region to identify selection designs that maximise selection outputs. Apart from providing a new approach to the problem of optimising selection system, the ASSSO also presents a new application of dynamic programming and branch-and-bound.
The ASSSO identified a number of alternative selection systems that are significant improvements to the practices currently used in the Burdekin region. Nevertheless, the purpose of this research was not to suggest that the intuitions and experiences of plant breeders can be replaced by the set of guidelines obtained using a computer simulation, but rather to validate the benefits of a joint venture between mathematicians and plant breeders.
|Item Type:||Thesis (PhD)|
Publications arising from this thesis are available from the Related URLs field. The publications are:
Calija, Vanja, Higgins, Andrew J., Jackson, Phillip A., Bielig, Leone M., and Coomans, Danny (2001) An operations research approach to the problem of the sugarcane selection. Annals of Operations Research, 108 (1-4). pp. 123-142.
Calija, V., Jackson, P., Higgins, A.J., Bielig, L., and Coomans, D. Simulating and optimising sugarcane selection. In: Operations research from theory to real life: proceedings of the 15th National Conference of the Australian Society for Operations Research Inc. ASOR Queensland Branch and ORSJ Hokkaido Chapter Joint Workshop. 15th National Conference of the Australian Society for Operations Research (ASOR Queensland Branch), 4-7 July 1999, Gold Coast, QLD, Australia, pp. 260-268.
Calija, V., Jackson, P., Higgins, A.J., Bielig, L., and Coomans, D. (1999) Simulating and optimising selection systems. Proceedings of the 11th Australian Plant Breeding Conference. 11th Australian Plant Breeding Conference, 19-23 April 1999, Adelaide, SA, Australia.
|Keywords:||Algorithm for the Sugarcane Selection Simulation Optimisation; ASSSO; Australian sugar industry; Burdekin plantations; Burdekin region; cane growers; cane growing; mathematical models; plant breeders; Queensland sugar industry; raw sugar; Saccharum; selection designs; selection strategies; selection systems; SSSM; sugar cane; sugar research; sugarcane breeding programs; sugarcane farmers; sugarcane farming; sugarcane farms; sugarcane genetics; sugarcane genotypes; Sugarcane Selection Simulation Model; sugarcane selection; sugarcane yield; sugarcane; sustainable cane growing|
|FoR Codes:||01 MATHEMATICAL SCIENCES > 0102 Applied Mathematics > 010202 Biological Mathematics @ 34%|
01 MATHEMATICAL SCIENCES > 0104 Statistics > 010402 Biostatistics @ 33%
06 BIOLOGICAL SCIENCES > 0604 Genetics > 060412 Quantitative Genetics (incl Disease and Trait Mapping Genetics) @ 33%
|SEO Codes:||82 PLANT PRODUCTION AND PLANT PRIMARY PRODUCTS > 8203 Industrial Crops > 820304 Sugar @ 100%|
|Deposited On:||05 Nov 2012 15:20|
|Last Modified:||03 Jan 2013 12:47|
Last 12 Months: 345
Repository Staff Only: item control page