A Hybrid Multi-Objective System Optimization Approach for Risk and Resilient Management in Multimodal Transportation Models
Loading...
Files
Date
2021
Authors
Emre, Beyza
Journal Title
Journal ISSN
Volume Title
Publisher
İzmir Ekonomi Üniversitesi
Open Access Color
OpenAIRE Downloads
OpenAIRE Views
Abstract
Günümüz taşımacılık operasyonlarında tek modlu taşımacılık yaygın olarak kullanılmasına rağmen taşımanın toplam maliyetini ve süresini azaltma olanağı sunması, aynı zamanda operasyonlar sırasında karşılaşılan risklere karşı esneklik sağlaması sebepleriyle çok modlu taşımacılığa yönelim artmaktadır. Farklı risk faktörleri altında taşımacılık firmalarının uğrayacakları etkileri azaltmak ve hangi risk faktörlerinin firma faaliyetleri üzerinde daha büyük etkiye sahip olduğunu saptayabilmek planlama faaliyetleri açısından oldukça önemli konumdadır. Bu çalışma kapsamında oluşturulan melez çözüm yaklaşımı çok amaçlı matematiksel model kullanılarak toplam taşımacılık zamanını, maliyetini ve karbon ayak izini minimize etmeyi amaçlarken, eş zamanlı olarak önceden tanımlanmış çok modlu taşımacılık faaliyetlerinde ortaya çıkan 7 adet risk faktörünü göz önüne alarak toplam risk etkisini minimize etmeye çalışmaktadır. Geliştirilen çok-amaçlı doğrusal karma tamsayı optimizasyon modeli epsilon-kısıt yöntemi ile çözülerek gerçek hayat veri seti ile test edilmiştir. Pareto-sonuç seti karar vericiler ile paylaşılmış ve karar vericilerin sürdürülebilir çok modlu taşımacılık faaliyetleri için alternatif risk faktörleri altında çok modlu taşımacılık faaliyetlerini planlaması ve geliştirmesi sağlanmıştır. Ayrıca, elde edilen farklı risk faktörleri altında elde edilen Pareto çözüm setleri (maliyet, zaman ve karbon ayak izi) t-dağıtılmış Stokastik Komşu Gömme (t-SNE) yöntemi kullanılarak boyut azaltımı yapıldıktan sonra makine öğrenimi sayesinde k-ortalama algoritması ile sektör kullanıcıları için alternatif risk kümeleri paylaşılmıştır.
Although unimodal transportation is widely used in today's transportation operations, tendency to multimodal transportation is increasing because it offers the opportunity to reduce the total cost and time of transportation and at the same time provides resilience against the risks encountered during operations. Reducing the effects that transportation companies can face under different risk factors and determining which risk factors have a greater impact on company activities takes an important place in terms of planning activities. The hybrid solution approach created within the scope of this study aims to minimize the total transportation time, cost, and carbon footprint by using a multi-objective mathematical model, while simultaneously trying to minimize the total risk impact by taking into account the 7 predefined risk factors that can occur in multi-modal transportation activities. Developed multi-objective linear mixed integer optimization model is solved by using epsilon-constraint method and tested with real life data set. Pareto-solution sets are shared with decision makers and decision makers have been enabled to plan and develop multimodal transportation activities under alternative risk factors for sustainable multimodal transportation activities. In addition, after performing size reduction to the obtained Pareto solution sets (cost, time, and carbon footprint) under obtained different risk factors, with the t-distributed Stochastic Neighbor Embedding (t-SNE) method, k-means algorithm is performed through machine learning and then alternative risk clusters are shared for users in sector.
Although unimodal transportation is widely used in today's transportation operations, tendency to multimodal transportation is increasing because it offers the opportunity to reduce the total cost and time of transportation and at the same time provides resilience against the risks encountered during operations. Reducing the effects that transportation companies can face under different risk factors and determining which risk factors have a greater impact on company activities takes an important place in terms of planning activities. The hybrid solution approach created within the scope of this study aims to minimize the total transportation time, cost, and carbon footprint by using a multi-objective mathematical model, while simultaneously trying to minimize the total risk impact by taking into account the 7 predefined risk factors that can occur in multi-modal transportation activities. Developed multi-objective linear mixed integer optimization model is solved by using epsilon-constraint method and tested with real life data set. Pareto-solution sets are shared with decision makers and decision makers have been enabled to plan and develop multimodal transportation activities under alternative risk factors for sustainable multimodal transportation activities. In addition, after performing size reduction to the obtained Pareto solution sets (cost, time, and carbon footprint) under obtained different risk factors, with the t-distributed Stochastic Neighbor Embedding (t-SNE) method, k-means algorithm is performed through machine learning and then alternative risk clusters are shared for users in sector.
Description
Keywords
Endüstri ve Endüstri Mühendisliği, Industrial and Industrial Engineering, Multimodal taşımacılık, Multimodal transport
Turkish CoHE Thesis Center URL
Fields of Science
Citation
WoS Q
N/A
Scopus Q
N/A
Source
Volume
Issue
Start Page
1
End Page
76
Collections
Sustainable Development Goals
8
DECENT WORK AND ECONOMIC GROWTH

9
INDUSTRY, INNOVATION AND INFRASTRUCTURE

11
SUSTAINABLE CITIES AND COMMUNITIES

13
CLIMATE ACTION

