Uncertainty Quantification In Population Models
Uncertainty in general can be in the form of numeric or non-numeric, where the latter is qualitative and the former quantitative in nature. In numerical quantities, uncertainty can be random in nature, in which case probability theory is appropriate, or it can be as a result of unclear informa...
Saved in:
| 主要作者: | |
|---|---|
| 格式: | Thesis |
| 語言: | English |
| 出版: |
2013
|
| 主題: | |
| 在線閱讀: | http://eprints.usm.my/45264/1/Almbrok%20Hussin%20Alsonosi%20Omar24.pdf |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 總結: | Uncertainty in general can be in the form of numeric or non-numeric, where the latter
is qualitative and the former quantitative in nature. In numerical quantities, uncertainty
can be random in nature, in which case probability theory is appropriate, or it can be
as a result of unclear information, whereby fuzzy set theory is useful.
Our concern will be on uncertainty in population models described by differential
equations and solved numerically. We select the predator-prey model and susceptible-
infected-recovered epidemic model to explore the uncertainty in the population models
through the initial states. For randomness, the normal distribution is selected to intro-
duce the uncertainty in the predator-prey model while we use the Beta distribution to
insert the uncertainty in the epidemic model. For the fuzzy approach, we consider a
triangular fuzzy number to treat the lack of information in the both models. |
|---|