A perturbation-based heuristic for the capacitated multisource Weber problem☆
Introduction
The continuous capacitated location–allocation problem with a fixed number of open facilities each with a constant capacity, which is also known as the capacitated multisource Weber problem, may be stated as follows: given the location of each fixed point (customer point), the demand at each fixed point, the transportation cost for the area of interest, the number of facilities to open, and the capacity of each of these facilities, the aim is to determine the location of each facility, and the allocation of customers to these open facilities (if more than one facilities are to be opened). Given
- n
the number of fixed points (or customer points)
- wj
demand or weight of customer j (j = 1, … , n)
location of customer j where , (j = 1, … , n)
- M
the number of facilities to be located
- b
fixed capacity of a facility where b ∈ N
Parameters
coordinates of facility i where
- xij
quantity assigned from facility i to customer j, i = 1, … , M, j = 1, … , n
Decision variables
The problem can be formulated as follows:where d(Xi, aj) represents the Euclidean distance between facility i and customer j.
(1) denotes the objective function which is the total transportation cost, (2) ensures that capacity constraints of the facilities are not violated, (3) guarantees that the demand of every customer is satisfied and (4) refers to non-negativity of the decision variables xij.
It can be noted that once the set of open facilities has been decided upon (e.g., if we fix the open facilities in the formulation), the resulting problem reduces to the usual Transportation Problem (TP) which can be solved optimally in polynomial time. In short, the problem is to find the best facility configuration.
In this study the value of b is set to where ⌈x⌉ is the smallest integer larger than or equal to x. Note that if we introduce a dummy customer with a 0 transportation cost and a demand equals to the remaining demand, e.g., . This customer is used only when solving the TP, but not at the location and the allocation stages.
Most of the work in the literature on the capacitated facility location concentrates on the discrete problem and the methods mainly used include dual-ascent based [11], cross decomposition method [13], constructive-type heuristic [10], [7] and Lagrangian relaxation heuristics [2], [1].
Other related work on the continuous location problem include Eben-Chaine et al. [8] who studied the case of capacitated facility location on a line, and Brimberg and Mladenovic [4], Brimberg et al. [3] and recently by Salhi and Gamal [12] who investigated the multisource Weber problem. To our knowledge, it is only Cooper [6] in the 1970s who attempted the capacitated continuous case. He presented exact and approximate methods for solving the transportation-location problem. The heuristic method described in this work is a modification of the alternating transportation-location method introduced in [6]. Here, the location method and the usual TP are alternately applied until there is no epsilon improvement in cost. We shall describe Cooper’s method [6] as this will be used as the foundation for our perturbation-based heuristic.
Firstly, M facilities are randomly chosen from the fixed points. Then, the TP using these M open facilities is solved to find the allocation for the capacitated problem. For each of the M independent set of allocations, containing ni fixed points where i = 1, … , M and , the new location of the facilities is found using the iterative procedure based on the Weiszfeld Equation which is given below:where the superscript k denotes the iteration number and represents all or a fraction of the jth customer demand that is assigned to facility i. Obviously as some customers may have their demand split because of the solution of the TP and hence some customers can be used more than once in Eq. (5) with their appropriate demand adjusted accordingly.
The location problem and the TP are alternately solved until there is no epsilon improvement in cost.
According to [6], ATL yields a convergent monotone nonincreasing sequence of values for the objective function. However, there is no guarantee that it will converge to the global minimum but the result, when not optimal, is found empirically to lie within ∼10%, and usually within 2–3%, of the optimal solution when tested on small instances.
The rest of the paper is structured as follows: in the next section, the modification on Cooper’s ATL is presented. Section 3 describes our perturbation-based heuristic and section Section 4 presents a neighbourhood reduction for solving the TP. Section 5 provides our computational results and our findings as well as some research issues are given in the last section.
Section snippets
A modified Cooper’s heuristic
In this section we present a scheme for generating initial solutions and implementations that consider the diversity of these solutions when addressing the capacitated problem. These ideas with a slight modification within Cooper’s algorithm are then combined to form our first heuristic which we refer to as the modified Cooper’s heuristic.
A perturbation-based scheme
A post optimisation procedure that attempts to improve the currently found solution by ATLAL for each Sd, d = 1, … , D is proposed. In this approach, the locations of the facilities found with the ATLAL heuristic are perturbed by taking into account the clustering of the borderline customers. These customers are defined as those which lie in between their nearest facility and their second nearest facility. In other words, the distance between the customers and their nearest and second nearest
Effect of neighbourhood reduction
It can be shown that a large amount of the total cpu time is consumed in solving the large number of TPs. Note that though the TP is solved in polynomial time, the use of such a procedure so many times renders the whole exercise computationally unattractive. There are a few ways on how to overcome this drawback. In this study, at each iteration, when solving the TP, we concentrate on a smaller portion of the original problem by considering a subset of facilities only. A smaller neighbourhood is
Computational results
The proposed heuristics are written in Fortran90 and run on Sun Enterprise Workstation 450 running Solaris 2.6. We used the four test problems given in the literature for the uncapacitated case, see [3]. These are the 50-fixed points, the 287-fixed points, 654-fixed points and the 1060-fixed points test problems. The weight of all customers is set to unity except for the 287-fixed point problem. The algorithms are applied to the test problems to solve for 2–25 open facilities for the 50-fixed
Conclusion and possible research issues
A perturbation-based heuristic is proposed to solve the capacitated continuous location–allocation problem which appears to have been scarcely investigated in the past. The heuristic uses the Furthest Distance Rule method to generate the initial starting locations for the uncapacitated problem though other rules were also tested. The uncapacitated problem is then solved using different starting locations for K times but only a sample of configurations are chosen as the starting locations for
Acknowledgements
The authors would like to thank the Malaysian Government for the sponsorship of the first author. We are also grateful to the referees for their constructive comments that improved both the content as well as the presentation of the paper.
References (14)
Lagrangean heuristic for location problems
European Journal of Operational Research
(1993)- et al.
ADD-heuristics starting procedures for capacitated plant location models
European Journal of Operational Research
(1985) Heuristics for the capacitated plant location model
European Journal of Operational Research
(1983)- et al.
Lagrangean heuristics applied to a variety of large capacitated plant location problems
Journal of The Operational Research Society
(1998) - et al.
Improvements and comparison of heuristics for solving the uncapacitated multisource Weber problem
Operations Research
(2000) - et al.
A variable neighbourhood algorithm for solving the continuous location–allocation problem
Studies in Locational Analysis
(1996) Heuristic methods for location–allocation problems
SIAM Review
(1964)
Cited by (34)
The location-allocation problem of drone base stations
2019, Computers and Operations ResearchData center supply chain configuration design: A two-stage decision approach
2019, Socio-Economic Planning SciencesCitation Excerpt :They used a combination of linear programming and simulated annealing technique to solve the linearized model. In a recent study, Tayeb et al. [25] presented two variants of hub-location model to identify optimal virtual machine placement decisions and used multi-commodity flow with variable aggregation and valid inequalities to solve the model. In the existing studies, however, selection of facility configuration based on demand characteristics has rarely been addressed.
The continuous single source location problem with capacity and zone-dependent fixed cost: Models and solution approaches
2017, European Journal of Operational ResearchThe capacitated multi-facility weber problem with polyhedral barriers: Efficient heuristic methods
2017, Computers and Industrial EngineeringA Two-echelon joint continuous-discrete location model
2017, European Journal of Operational ResearchCitation Excerpt :In these problems, there is an upper bound on the total customer demand that can be satisfied by a facility. See, for instance, Zainuddin and Salhi (2007) and Martino, Salhi, and Nagy (2009). The implicit assumption in the uncapacitated or capacitated MWP is that any facility so located has sufficient supply of commodities to cater to demand points assigned to it.
Transportation-location problem with unknown number of facilities
2017, Computers and Industrial EngineeringCitation Excerpt :In this study we extend the first heuristic in Cooper (1972) to the case with unknown number of new facilities (i.e., the TLUF) and incorporate stochasticity to the heuristic in order to escape local minimums. Cooper (1976), Murtagh and Niwattisyawong (1982), Zainuddin and Salhi (2007), Mohammadi, Malek, and Alesheikh (2010), and Luis, Salhi, and Nagy (2011) also propose heuristics for the TLP based on the Cooper’s decomposition. Other authors such as Sherali and Tuncbilek (1992), Sherali, Ramachandran, and Kim (1994), Sherali, Al-Loughani, and Subramanian (2002), Aras, Altinel, and Orbay (2007), Aras, Orbay, and Altinel (2008), Boyaci, Altinel, and Aras (2013), Akyüz, Altinel, and Öncan (2014) study the TLP considering alternative Minkowski distances.
- ☆
This research was conducted when both authors were at the University of Birmingham, UK.