Irrigation Scheduling with Integer Programming
Irrigation Scheduling with Integer Programming
Water delivery arrangements in irrigation systems range from completely fixed where duration, discharge and start time are set once and seldom or never change to on-demand, where users can take water whenever they need it in whatever quantity desired. In between these two extremes lies a range of arranged-demand systems where duration, discharge and/or start time are flexible to a certain degree but are agreed upon before the irrigation event takes place.
This thesis shows how arranged-demand scheduling can be interpreted as a single or multi-machine scheduling problem. Mixed integer linear programming is used to develop a series of models that allow scheduling of irrigation turns in a wide range of irrigation systems. Given that duration and target time of the irrigation turns are specified by the users, the objective of all models is to find a schedule such that the difference between target start time and scheduled start time are as small as possible.
Two types of models are presented that reflect different management options at tertiary level. Contiguous scheduling schedules irrigation turns back-to-back so that operational spillage and/or gate operations are minimised. In non-contiguous schedules idle time can be inserted in between irrigation turns, which allows greater opportunity to match target start times with scheduled start times. Travel times play an important role in open channel irrigation systems and need to be taken into account when determining irrigation schedules. In pressurised systems on the other hand, travel times are non-existent. Models that incorporate travel times as well as models without travel times are presented to demonstrate scheduling of irrigation turns in both open channel and pressurised irrigation systems.
University of Southampton
de Vries, Tonny Tessa
338ad4ac-c215-4232-a42c-5b99cef9d397
2003
de Vries, Tonny Tessa
338ad4ac-c215-4232-a42c-5b99cef9d397
de Vries, Tonny Tessa
(2003)
Irrigation Scheduling with Integer Programming.
University of Southampton, Doctoral Thesis.
Record type:
Thesis
(Doctoral)
Abstract
Water delivery arrangements in irrigation systems range from completely fixed where duration, discharge and start time are set once and seldom or never change to on-demand, where users can take water whenever they need it in whatever quantity desired. In between these two extremes lies a range of arranged-demand systems where duration, discharge and/or start time are flexible to a certain degree but are agreed upon before the irrigation event takes place.
This thesis shows how arranged-demand scheduling can be interpreted as a single or multi-machine scheduling problem. Mixed integer linear programming is used to develop a series of models that allow scheduling of irrigation turns in a wide range of irrigation systems. Given that duration and target time of the irrigation turns are specified by the users, the objective of all models is to find a schedule such that the difference between target start time and scheduled start time are as small as possible.
Two types of models are presented that reflect different management options at tertiary level. Contiguous scheduling schedules irrigation turns back-to-back so that operational spillage and/or gate operations are minimised. In non-contiguous schedules idle time can be inserted in between irrigation turns, which allows greater opportunity to match target start times with scheduled start times. Travel times play an important role in open channel irrigation systems and need to be taken into account when determining irrigation schedules. In pressurised systems on the other hand, travel times are non-existent. Models that incorporate travel times as well as models without travel times are presented to demonstrate scheduling of irrigation turns in both open channel and pressurised irrigation systems.
Text
903635.pdf
- Version of Record
More information
Published date: 2003
Identifiers
Local EPrints ID: 464974
URI: http://eprints.soton.ac.uk/id/eprint/464974
PURE UUID: efba579b-5cd2-4b8a-a1d3-b7be9b436cd9
Catalogue record
Date deposited: 05 Jul 2022 00:14
Last modified: 16 Mar 2024 19:51
Export record
Contributors
Author:
Tonny Tessa de Vries
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics