Loading...
Search
Search in this resource
sort by
Send to Friend
 Digital Object List
 Bookmark
 Review List
  1. Simple search
  2. Search result
  3. بهینه‌سازی طراحی شبکه زنجیره تامین خون با ادغام روش‌های یادگیری‌ ماشین و تصمیم‌گیری چندمعیاره
بهینه‌سازی طراحی شبکه زنجیره تامین خون با ادغام روش‌های یادگیری‌ ماشین و تصمیم‌گیری چندمعیاره
113 viewed

بهینه‌سازی طراحی شبکه زنجیره تامین خون با ادغام روش‌های یادگیری‌ ماشین و تصمیم‌گیری چندمعیاره

عموجعفری، امیر حسین Amou Jafari, Amir Hossein

Optimization Blood Supply Chain Network Design using Machine Learning and Multi-Criteria-Decision-Making

Amou Jafari, Amir Hossein | 2026

113 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 58771 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Akbari Jokar, Mohammad Reza
  7. Abstract:
  8. Blood supply chain network design, owing to the critical nature and perishability of blood products and their strong dependence on spatial–temporal conditions, is one of the most complex logistics networks in healthcare. Despite a growing body of literature, there remain notable gaps in the integrated treatment of data-driven spatial analysis, location scoring, and stochastic network modeling. This study proposes a unified approach for the design of an urban blood supply chain network. First, an improved spatial clustering scheme is employed to partition urban areas into compact and balanced clusters that are homogeneous in terms of demand patterns and spatial characteristics. Second, for candidate sites of blood collection centers, a data-driven scoring mechanism is developed which, depending on data availability and quality, relies either on a neural network–based model or, under data limitations, on a multicriteria decision-making framework combining MAUT and TOPSIS, thereby translating locational attractiveness, accessibility, and alignment with urban and health criteria into quantitative scores. These outputs serve as inputs to a two-stage stochastic mixed-integer programming model, in which strategic-level decisions determine the location and type of facilities (permanent or temporary), while tactical–operational decisions specify flows, allocations, and service levels under multiple scenarios of uncertain supply and demand. A case study based on data from the city of Tehran shows that the improved clustering alone yields approximately a 5–10% reduction in total unmet demand compared with baseline clustering structures, while the proposed location-scoring mechanism delivers an additional 4–6% reduction in unmet demand in network expansion scenarios. Overall, the proposed data-driven framework substantially enhances both the efficiency and the spatial equity of blood supply chain network design at the urban scale
  9. Keywords:
  10. Two Stage Stochastic Programming ; Blood Supply Chain Optimization ; Multicriteria Decision Making ; Health Care System ; Machine Learning ; Improved Clustering ; Blood Supply Chain Netwrok Desgin

 Digital Object List

 Bookmark

No TOC