| Author | Saisunee Budhakooncharoen |
| Call Number | AIT Diss. no. WA-90-01 |
| Subject(s) | Reservoirs--Mathematical models
|
| Note | A dissertation submitted in partial fulfilment of the
requirements for the degree of Doctor of Engineering, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Dissertation ; no. WA-90-01 |
| Abstract | An interactive Multi-Objective Decision Making (MODM) model
is developed to derive the non- dominated operational policies of
reservoirs in MODM environment. The model relies basically on
the interlinkage of two major parts, an alternative generating
part and a subsequent evaluation part.
To derive a set of non- dominated operating rules in . the
alternative generating part, a new method, the Search Beam
Model is developed. It is based on the systematic modified
Fibonacci Search or Golden Cut procedures to locate the nondominated policy a long a line (Search Beam). The direction of
this line in the objective space is determined by the relative
preferences the Decision Maker (DM) assigns to the different
objectives. The reservoir operational policy is derived by an
appropriate method. In this study Incremental Dynamic
Programming (IDP) is applied for pre-specified periods.
The subsequent evaluation part plays an important role to
check whether the solution obtained in the alternative generating
part is satisficing. This is accomplished by introducing to the
DM the results of the reliability analysis of the reservoir
operational policy obtained from the Search Beam Model. The
relevant reliability indicators or performance indices (Pl) are
determined in such a manner that the DM can obtain a wealth of
information concerning each alternative policy which may be a
candidate for being the satisficing solution. If the solution is
not satisficing, the DM is requested to express new relative
preferences on each objective. Then , new non-dominated
operation policy will be derived in the alternative generating
part and the whole cycle is repeated. This iterative process
will be carried out until the DM is satisfied with the solution.
The search will be automatically terminated upon reaching t he
satisfaction of the DM.
To assist the location of the 'most preferred ' (satisficing)
alternative, ELECTRE I was applied to reduce the discrete
Pareto set by detecting the outranking alternatives. The
analysis was based on both objectives and the reliability
assessment. The DM will have to select the most preferred
alternative among the outranking solutions.
The practical viability of the model is shown by applying it
to two case studies, t wo single unit water resources systems , the Kariba Reservoir in Central Africa and the Ubol Ratana
Reservoir in Thailand. The non-dominated operational policies of
the former one are derived by assuming objectives to maximize
( H)
ABSTRACT (Cont'd)
water surface e l evat ion and hydropower output, while the
objective criteria of the latter one are to maximize irrigation
water supply and hydropower generation .
|
| Year | 1990 |
| Corresponding Series Added Entry | Asian Institute of Technology. Dissertation ; no. WA-90-01 |
| Type | Dissertation |
| School | School of Engineering and Technology |
| Department | Other Field of Studies (No Department) |
| Academic Program/FoS | Water Resources Research Engineering (WA) |
| Chairperson(s) | Harboe, Ricardo ;Bogardi, Janos J.
; |
| Examination Committee(s) | Gupta, Ashim Das ;Paudyal, Guna N. ;Tabucanon, Mario T. ;McBean, Edward ; |
| Scholarship Donor(s) | Deutsche Gesellschaft fur Technische
Zusarnmenarbeit (GTZ), GmbH
Federal Republic of Germany ; |
| Degree | Thesis (Ph.D.) - Asian Institute of Technology, 1990 |