Deterministic and stochastic inventory routing problems with backorders using artificial bee colony / Huda Zuhrah Ab Halim

Huda Zuhrah , Ab Halim (2019) Deterministic and stochastic inventory routing problems with backorders using artificial bee colony / Huda Zuhrah Ab Halim. PhD thesis, Universiti Malaya.

[img] PDF (The Candidate's Agreement)
Restricted to Repository staff only

Download (212Kb)
    PDF (Thesis PhD)
    Download (3759Kb) | Preview


      This thesis is devoted to solving the inventory routing problem (IRP) and its variants. It is well known that the IRP is an important component in the implementation of Vendor Managed Inventory. The IRP comprises the coordination of two components: inventory management and vehicle routing problem. Details of the components define the variation of the IRP. Four different variants of IRP are studied in this thesis. The first variant is a many-to-one distribution network, consisting of a depot, an assembly plant, and geographically dispersed suppliers where a capacitated homogeneous vehicle, housed at the depot delivers a distinct product from the suppliers to fulfill the deterministic demand specified by the assembly plant. The inventory holding cost is assumed to be product specific and only incurred at the assembly plant. An Artificial Bee Colony (ABC) algorithm is proposed for the problem. The performance of ABC is evaluated on existing datasets and compared with Scatter Search (SS) and Genetic Algorithm (GA). The statistical analysis carried out shows that the ABC, SS and GA are significantly different with 95% confidence level with ABC performing significantly better compared to SS and GA. An enhanced ABC is also developed which performs better in terms of quality of the solutions. IRP with backordering (IRPB) represents the second variant and it defines the condition where unsatisfied demand is delayed and fulfilled in future periods. The network of IRPB consists of a supplier and geographically scattered customers, where a set of vehicles performs the delivery to fulfill customer’s demand. Two ABC algorithms are proposed which embed two inventory updating mechanisms; random exchange and guided exchange. Results of both ABCs are compared with existing literature and bounds found by CPLEX. The statistical analysis result shows that all the algorithms are significantly different with 95% confidence level. The third variant investigated is IRP with stochastic demand. The main characteristic of the demand where the demand is known in a probabilistic sense and the demand is gradually revealed at the end of each period (dynamic). The problem is known as dynamic and stochastic inventory routing problem (DSIRP) and the distribution network considered consists of a supplier and a set of retailers. An order-up-to level inventory policy is applied, and each unit of positive inventory incurs a holding cost while a penalty is incurred for each negative inventory level. The transportation of the product is handled by a third party. The DSIRP is modeled as stochastic dynamic programming and solved using a matheuristic, enhanced hybrid rollout algorithm. The enhanced algorithm embeds additional controls generated using ABC. The DSIRP is then extended to handle backorder decisions (DSIRPB) which is the fourth variant of IRP studied. A new MILP for DSIRPB is formulated and used within the algorithm. The DSIRPB is modeled as stochastic dynamic programming and solved using hybrid rollout algorithm. Both DSIRP and DSIRPB are evaluated on 60 instances. Analysis of controls, the number of visits, quantity delivery and backorders are carried out to observe the patterns.

      Item Type: Thesis (PhD)
      Additional Information: Thesis (PhD) - Faculty of Science, Universiti Malaya, 2019.
      Uncontrolled Keywords: Inventory routing problem; Artificial Bee Colony; backordering; Stochastic IRP; Matheuristic
      Subjects: Q Science > Q Science (General)
      Divisions: Faculty of Science
      Depositing User: Mr Mohd Safri Tahir
      Date Deposited: 10 Mar 2021 03:32
      Last Modified: 06 Jan 2022 06:16

      Actions (For repository staff only : Login required)

      View Item