Back to search

University of Thessaly (UTH)

Online αλγόριθμοι δρομολόγησης για στοχαστικά δίκτυα σε πραγματικό χρόνο

Abstract

dc:description

This thesis investigates the general truck dispatching problem known as the “Vehicle Routing Problem” (VRP). It is one of the most frequently studied problems in the field of applied mathematics and perhaps the most renowned problem of operational research in the transport sector. It is defined as the determination of the least cost distribution plans of a fleet of vehicles to a network of customers, while concurrently ensuring indicative systematic constraints. Applications range from fleet management and product distribution to bus school routing and waste collection.The primary goals of this study are to analyze the VRP under uncertainty (e.g. demand variation) and embed stochasticity in the development of the corresponding mathematical models and algorithms. When an optimization problem is solved under uncertainty the goal is to determine the policies that minimize the cost of initial decisions and the cost of feasible recourse actions. Taking into account the necessity of robust assessments of revealed choices in the optimization model under uncertainty, it is crucial to clearly formulate (mathematical modeling) all possible future decisions – outcomes and to incorporate their effect into a priori decisions. The initial methodological pillars that are settled for the investigation of the VRP under uncertainty in this study are stochastic programming and on line optimization. The first part of the research thesis focused on investigating exact algorithms for the VRP. Extended literature review is performed focusing on alternative model formulation.With focus to freight transportation, the aim is to identify a route plan for delivering products fulfilling demand and minimizing transportation cost. When the demand of products varies the problem of determining route plans from deterministic becomes stochastic. Stochastic demand is directly related to product absorption rate, level of inventory control as well as their correlation. Concentrating interest to stochasticity of demand in coherent to factors affected (capacity of inventories, product absorption rate) the dissertation is studying the collaboration of transportation management and inventory control, known as “Inventory routing problem”.In the second part of this thesis an extended review of the relevant literature of inventory routing problem is performed. Excluding the basic assumption of all previous proposed models it was proved that a new policy called proactive deliveries was powerful in the framework of deterministic modeling. Aiming to solve the inventory routing problem under uncertainty and securing flexibility, a recourse action of the transshipment was proposed in the framework of two stage stochastic programming model. An innovative model was developed of two stage recourse action for the stochastic inventory routing problem. An L – Shaped decomposition method was adopted to solve the problem exactly. The proposed model was tested using a modification of known benchmark instances (Arhetti et al). Conclusively, the decision of formulating the inventory routing problem under uncertainty in the framework of stochastic programming of two stage recourse action model gave us the opportunities to clearly state the tradeoff between transshipment cost in combination with proactive distribution plans and increased inventory levels.

Degree

thesis:*
Grantor dc:publisher
University of Thessaly (UTH)
Year dc:date
2015

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Chrysochoou, Evangelia
  • Χρυσοχόου, Ευαγγελία

Subjects

dc:subject × 10

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/36374

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Chrysochoou, Evangelia; Χρυσοχόου, Ευαγγελία. Online αλγόριθμοι δρομολόγησης για στοχαστικά δίκτυα σε πραγματικό χρόνο. University of Thessaly (UTH), 2015. http://hdl.handle.net/10442/hedi/36374