The Allocation of Funds Based on the Optimization Model
Konferenz: ICMLCA 2021 - 2nd International Conference on Machine Learning and Computer Application
17.12.2021 - 19.12.2021 in Shenyang, China
Tagungsband: ICMLCA 2021
Seiten: 8Sprache: EnglischTyp: PDF
Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt
Autoren:
Shen, Peiran (20 Wukesong Rd, Haidian District, Beijing, China)
You, Zhibin (70 Yongding Rd, Haidian District, Beijing, China)
Wang, Haotian (Beiwa Road, Haidian District, Beijing, China)
Liu, Guoxi (Zengguang Rd, Haidian District, Beijing, Chnina)
Inhalt:
As a variety of imperiled plants conservation projects lack sufficient funding to support their execution, establishing an efficacious fund-raising plan has become a critical issue. Our team decided to help the Florida Rare Plant Conservation Endowment (FRPCE) make the optimal fund-raising plan using mathematical models. To demonstrate our methodology, we utilized 48 endangered species in Florida as an example. In the database of the 48 plants, we found the data lacked a distinctive distribution pattern: most data were close to each other with a distant outlier. To address the challenge, our team constructed the optimal fundraising model combined with the 0-1 programming model. In addition, considering discount rate and changes of duration in the model, the team performed sensitivity analysis of the optimal fundraising schedule. The result indicates that the annual cost for conserving imperiled plants is sensitive to the changes of the overall conservation time period. Furthermore, the team applied an evaluation model to weigh every imperiled species with Analytic Hierarchy Process. Then, we applied the weight to calculate and rank the final weighted score of each project to acquire the priority sequence for conservation. Including all 48 species, the team determined the total length of conservation projects to be 26 years. Moreover, we ran computer programs in Lingo to minimize duration and standard deviation. The team also considered rate of depreciation and the influence of extreme data, shortening the overall time length to 24 years. Through modelling, we received a promising result that complies with our expectation.