A hybrid tabu search based heuristic for the periodic distribution inventory problem with perishable goods

Ali Diabat, Tarek Abdallah, Tung Le

    Research output: Contribution to journalArticle

    Abstract

    Most of the research on integrated inventory and routing problems ignores the case when products are perishable. However, considering the integrated problem with perishable goods is crucial since any discrepancy between the routing and inventory cost can double down the risk of higher obsolescence costs due to the limited shelf-life of the products. In this paper, we consider a distribution problem involving a depot, a set of customers and a homogeneous fleet of capacitated vehicles. Perishable goods are transported from the depot to customers in such a way that out-of-stock situations never occur. The objective is to simultaneously determine the inventory and routing decisions over a given time horizon such that total transportation cost is minimized. We present a new “arc-based formulation” for the problem which is deemed more suitable for our new tabu search based approach for solving the problem. We perform a thorough sensitivity analysis for each of the tabu search parameters individually and use the obtained gaps to fine-tune the parameter values that are used in solving larger sized instances of the problem. We solve different sizes of randomly generated instances and compare the results obtained using the tabu search algorithm to those obtained by solving the problem using CPLEX and a recently published column generation algorithm. Our computational experiments demonstrate that the tabu search algorithm is capable of obtaining a near-optimal solution in less computational time than the time required to solve the problem to optimality using CPLEX, and outperforms the column generation algorithm for solving the “path flow formulation” of the problem in terms of solution quality in almost all of the considered instances.

    Original languageEnglish (US)
    Pages (from-to)373-398
    Number of pages26
    JournalAnnals of Operations Research
    Volume242
    Issue number2
    DOIs
    StatePublished - Jul 1 2016

    Fingerprint

    Tabu search
    Heuristics
    Perishable goods
    Routing
    Integrated
    Column generation
    Optimality
    Transportation costs
    Sensitivity analysis
    Obsolescence
    Costs
    Optimal solution
    Shelf life
    Time horizon
    Discrepancy
    Experiment
    Inventory cost

    Keywords

    • Integer programming
    • Inventory-routing
    • Perishable goods
    • Tabu search

    ASJC Scopus subject areas

    • Decision Sciences(all)
    • Management Science and Operations Research

    Cite this

    A hybrid tabu search based heuristic for the periodic distribution inventory problem with perishable goods. / Diabat, Ali; Abdallah, Tarek; Le, Tung.

    In: Annals of Operations Research, Vol. 242, No. 2, 01.07.2016, p. 373-398.

    Research output: Contribution to journalArticle

    @article{4e4843a89c004dcaa00e5febe0905af9,
    title = "A hybrid tabu search based heuristic for the periodic distribution inventory problem with perishable goods",
    abstract = "Most of the research on integrated inventory and routing problems ignores the case when products are perishable. However, considering the integrated problem with perishable goods is crucial since any discrepancy between the routing and inventory cost can double down the risk of higher obsolescence costs due to the limited shelf-life of the products. In this paper, we consider a distribution problem involving a depot, a set of customers and a homogeneous fleet of capacitated vehicles. Perishable goods are transported from the depot to customers in such a way that out-of-stock situations never occur. The objective is to simultaneously determine the inventory and routing decisions over a given time horizon such that total transportation cost is minimized. We present a new “arc-based formulation” for the problem which is deemed more suitable for our new tabu search based approach for solving the problem. We perform a thorough sensitivity analysis for each of the tabu search parameters individually and use the obtained gaps to fine-tune the parameter values that are used in solving larger sized instances of the problem. We solve different sizes of randomly generated instances and compare the results obtained using the tabu search algorithm to those obtained by solving the problem using CPLEX and a recently published column generation algorithm. Our computational experiments demonstrate that the tabu search algorithm is capable of obtaining a near-optimal solution in less computational time than the time required to solve the problem to optimality using CPLEX, and outperforms the column generation algorithm for solving the “path flow formulation” of the problem in terms of solution quality in almost all of the considered instances.",
    keywords = "Integer programming, Inventory-routing, Perishable goods, Tabu search",
    author = "Ali Diabat and Tarek Abdallah and Tung Le",
    year = "2016",
    month = "7",
    day = "1",
    doi = "10.1007/s10479-014-1640-4",
    language = "English (US)",
    volume = "242",
    pages = "373--398",
    journal = "Annals of Operations Research",
    issn = "0254-5330",
    publisher = "Springer Netherlands",
    number = "2",

    }

    TY - JOUR

    T1 - A hybrid tabu search based heuristic for the periodic distribution inventory problem with perishable goods

    AU - Diabat, Ali

    AU - Abdallah, Tarek

    AU - Le, Tung

    PY - 2016/7/1

    Y1 - 2016/7/1

    N2 - Most of the research on integrated inventory and routing problems ignores the case when products are perishable. However, considering the integrated problem with perishable goods is crucial since any discrepancy between the routing and inventory cost can double down the risk of higher obsolescence costs due to the limited shelf-life of the products. In this paper, we consider a distribution problem involving a depot, a set of customers and a homogeneous fleet of capacitated vehicles. Perishable goods are transported from the depot to customers in such a way that out-of-stock situations never occur. The objective is to simultaneously determine the inventory and routing decisions over a given time horizon such that total transportation cost is minimized. We present a new “arc-based formulation” for the problem which is deemed more suitable for our new tabu search based approach for solving the problem. We perform a thorough sensitivity analysis for each of the tabu search parameters individually and use the obtained gaps to fine-tune the parameter values that are used in solving larger sized instances of the problem. We solve different sizes of randomly generated instances and compare the results obtained using the tabu search algorithm to those obtained by solving the problem using CPLEX and a recently published column generation algorithm. Our computational experiments demonstrate that the tabu search algorithm is capable of obtaining a near-optimal solution in less computational time than the time required to solve the problem to optimality using CPLEX, and outperforms the column generation algorithm for solving the “path flow formulation” of the problem in terms of solution quality in almost all of the considered instances.

    AB - Most of the research on integrated inventory and routing problems ignores the case when products are perishable. However, considering the integrated problem with perishable goods is crucial since any discrepancy between the routing and inventory cost can double down the risk of higher obsolescence costs due to the limited shelf-life of the products. In this paper, we consider a distribution problem involving a depot, a set of customers and a homogeneous fleet of capacitated vehicles. Perishable goods are transported from the depot to customers in such a way that out-of-stock situations never occur. The objective is to simultaneously determine the inventory and routing decisions over a given time horizon such that total transportation cost is minimized. We present a new “arc-based formulation” for the problem which is deemed more suitable for our new tabu search based approach for solving the problem. We perform a thorough sensitivity analysis for each of the tabu search parameters individually and use the obtained gaps to fine-tune the parameter values that are used in solving larger sized instances of the problem. We solve different sizes of randomly generated instances and compare the results obtained using the tabu search algorithm to those obtained by solving the problem using CPLEX and a recently published column generation algorithm. Our computational experiments demonstrate that the tabu search algorithm is capable of obtaining a near-optimal solution in less computational time than the time required to solve the problem to optimality using CPLEX, and outperforms the column generation algorithm for solving the “path flow formulation” of the problem in terms of solution quality in almost all of the considered instances.

    KW - Integer programming

    KW - Inventory-routing

    KW - Perishable goods

    KW - Tabu search

    UR - http://www.scopus.com/inward/record.url?scp=84901868760&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84901868760&partnerID=8YFLogxK

    U2 - 10.1007/s10479-014-1640-4

    DO - 10.1007/s10479-014-1640-4

    M3 - Article

    VL - 242

    SP - 373

    EP - 398

    JO - Annals of Operations Research

    JF - Annals of Operations Research

    SN - 0254-5330

    IS - 2

    ER -