Abstract
The Multi-commodity Capacitated Multi-facility Weber Problem (MCMWP) is concerned with locating I-capacitated facilities in the plane in order to satisfy the demands of J customers for K commodities so that the total transportation cost is minimized. We propose a Lagrangean relaxation scheme and a subgradient-like algorithm based on the relaxation of the capacity and commodity bundle constraints. The resulting subproblem is a variant of the well-known Multi-facility Weber Problem and it can be solved by using column generation and branch-and-price on an equivalent set covering formulation, which is accurate but extremely inefficient. Therefore, we devise different strategies to increase the efficiency. They mainly benefit from the effective usage of the lower and upper bounds on the optimal value of the Lagrangean subproblem. On the basis of extensive computational tests, we can say that they increase the efficiency considerably and result in accurate Lagrangean heuristics.
Similar content being viewed by others
References
Akyüz MH, Öncan T and Altınel İK (2009). Efficient lower and upper bounds for the multi-commodity capacitated multi-facility weber problem with rectilinear distances. In: Voss S, Pahl J and Schwarze S (eds). Logistik Management Systeme, Methoden, Integration. Physica-Verlag: Heidelberg, pp 229–246.
Akyüz MH, Öncan T and Altınel İK (2010a). The multi-commodity capacitated multi-facility Weber problem: Heuristics and confidence intervals. IIE Trans 42: 825–841.
Akyüz MH, Öncan T and Altınel İK (2010b). Efficient approximate solution methods for the multi-commodity capacitated multi-facility Weber problem. Research Paper No FBE-IE-02/2010-03, Institute for Graduate Studies in Science and Engineering, Boğaziçi University.
Al-Loughani L (1997). Algorithmic approaches for solving the Euclidean distance location-allocation problems. PhD Thesis, Virginia Polytechnic Institute and State University, USA.
Alpaydin E, Altınel İK and Aras N (1996). Parametric distance functions versus nonparametric neural networks for estimating road travel distances. Eur J Opl Res 93 (2): 230–243.
Aras N, Altınel İK and Orbay M (2007). New heuristic methods for the capacitated multi-facility Weber problem. Nav Res Log 54: 21–32.
Aras N, Orbay M and Altınel İK (2008). Efficient heuristics for the rectilinear distance capacitated multi-facility Weber problem. J Opl Res Soc 59: 64–79.
Brimberg J and Love RF (1993). Global convergence of a generalized iterative procedure for the minisum location problem with lp distances. Opns Res 41: 1153–1163.
Brimberg J, Chen R and Chen D (1998). Accelerating convergence in the Fermat-Weber location problem. Opns Res 22: 151–157.
Brimberg J, Walker JH and Love RF (2007). Estimation of travel distances with the weighted lp norm: Some empirical results. J. Transp Geogr 15 (1): 62–72.
Chen P, Hansen P, Jaumard B and Tuy H (1998). Solution of the multisource Weber and conditional Weber problems by D.C. programming. Opns Res 46 (4): 548–562.
Cooper L (1964). Heuristic methods for location-allocation problems. SIAM Rev 6: 37–53.
Cooper L (1972). The transportation-location problem. Opns Res 20: 94–108.
Drezner Z (1984). The planar two-center and two-median problems. Trans Sci 18 (4): 352–361.
Drezner Z, Mehrez A and Wesolowsky GO (1991). The facility location problem with limited distances. Trans Sci 25 (3): 183–187.
Fisher RA and Tippett LHC (1928). Limiting forms of the frequency distribution of the largest or smallest member of a sample. Proc Cambridge Philosoph Soc 24: 180–190.
Frenk JBG, Melo MT and Zhang S (1994). A Weiszfeld method for acgeneralized l p distance minisum location model in continuous space. Location Sci 2: 111–117.
Hansen P, Perreur J and Thisse JF (1980). Location theory, dominance and convexity: Some further results. Opns Res 28 (5): 1241–1250.
Hansen P, Mladenovic N and Taillard E (1998). Heuristic solution of the multisource Weber problem as a p-median problem. Opns Res Lett 22: 55–62.
Idrissi H, Loridan P and Michelot C (1988). Approximation of solutions for location problems. J Optim Theory Appl 56: 127–143.
Klingman D, Napier A and Stutz J (1974). NETGEN: A program for generating large scale capacitated assignment, transportation, and minimum cost flow network problems. Mngt Sci 20: 814–821.
Krau S (1997). Extensions du probleme de Weber. PhD Thesis, University of Montreal.
Luis M, Salhi S and Nagy G (2009). Region-rejection based heuristics for the capacitated multi-source Weber problem. Comput Opns Res 36: 2007–2017.
Megiddo N and Supowit KJ (1984). On the complexity of some common geometric location problems. SIAM J Comput 13 (1): 182–196.
Polyak BT (1967). A general method of solving extremum problems. Soviet Math Dokl 8: 593–597.
Polyak BT (1969). Minimization of unsmooth functionals. USSR Comput Math Math Phys 9 (3): 14–29.
Righini G and Zaniboni L (2007). A branch-and-price algorithm for the multi-source Weber problem. Int J Opns Res 2 (2): 188–207.
Rosing KE (1992). An optimal method for solving the (generalized) multi-Weber problem. Eur J Opl Res 58: 414–426.
Rosing KE and ReVelle CS (1997). Heuristic concentration: Two stage solution construction. Eur J Opl Res 97: 75–86.
Selim S (1979). Biconvex programming and deterministic and stochastic location allocation problems. PhD Thesis, Georgia Institute of Technology, USA.
Sherali HD and Adams WP (1999). A Reformulation-linearization Technique for Solving Discrete and Continuous Nonconvex Problems. Kluwer Academic Publishers: the Netherlands.
Sherali HD and Tunçbilek CH (1992). A squared-Euclidean distance location-allocation problem. Nav Res Log 39 (4): 447–469.
Sherali HD, Ramachandran S and Kim SI (1994). A localization and reformulation discrete programming approach for the rectilinear distance location-allocation problems. Disc Appl Math 49 (1–3): 357–378.
Sherali HD, Al-Loughani I and Subramanian S (2002). Global optimization procedures for the capacitated Euclidean and ℓp distance multifacility location-allocation problems. Opns Res 50 (3): 433–448.
Thisse JF, Ward JE and Wendell RE (1984). Some properties of location problems with block and round norms. Opns Res 32 (6): 1309–1327.
Ward JE and Wendell RE (1980). A new norm for measuring distance which yields linear location problems. Opns Res 28 (3): 836–844.
Ward JE and Wendell RE (1985). Using block norms for location modeling. Opns Res 33 (5): 1074–1090.
Weiszfeld E (1937). Sur le point lequel la somme des distances de n points donnes est minimum. Tôhoku Math J 43: 355–386.
Wendell RE and Hurter AP (1973). Location theory, dominance and convexity. Opns Res 21: 314–320.
Witzgall C (1964). Optimal location of central facility: Mathematical models and concepts. National Bureau of Standards, Report No. 8388.
Zainuddin ZM and Salhi S (2007). A perturbation-based heuristic for the capacitated multisource Weber problem. Eur J Opl Res 179: 1194–1207.
Acknowledgements
We thank the two anonymous referees for their valuable comments, which have helped improve this article. This research was supported by the Turkish Scientific and Technological Research Council (TÜBİTAK) Research Grant No: 107M462, and Galatasaray University Scientific Research Projects Grant No: 07.402.014, 10.402.001 and 10.402.019. The first author acknowledges the partial support of National Graduate Scholarship Program for PhD Students awarded by TÜBİTAK.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
A.1. Subgradient-like algorithm
Algorithm 1 SL Algorithm
-
1
Initialization: Set β ik ≔ 0, μ ij ≔0 for all i , j and k , π ≔2, z LB best ≔−∞, z UB best ≔∞, Repeat steps 2–5 until the algorithm converges,
-
2
Find heuristic bound z H ( β , μ ) with Lagrange multipliers β ik and μ ij . If z H ( β , μ )> z UB best , then find a valid lower bound z LB with β ik and μ ij on the optimum z * and update z LB best if necessary (ie, z LB best ≔max { z LB , z LB best }). Otherwise, set z LB = z H ( β , μ )
-
3
Find an upper bound z UB on z * If z UB < z UB best , then set z UB best := z UB ,
-
4
Update multipliers by setting and where and Set π ≔ π/2 if z LB did not improve within the last 30 iterations,
-
5
Find a final valid lower bound z LB with the best β ik and μ ij on z * , update z LB best as necessary and output z LB best and z UB best .
The current lower bound values z LB computed by the SL algorithm can be either an optimum solution of the LR subproblem z LR *(β, μ) or a lower bound satisfying z LB ⩽z LR *(β, μ) for given multipliers β and μ. The upper bounds are determined by an Alternate Location Allocation (ALA)-type heuristic (ie, Cooper, 1964, 1972) which is initialized with the allocation values w ijk =α ijk × q jk obtained from the solution of the LR subproblem.
When using heuristic solution values z H (β, μ), we set z LB =z H (β, μ) through the steps of the algorithm. In case z H (β, μ)>z UB best occurs at some step of the algorithm, we switch back to previous multipliers and solve one of the recovery algorithms (ie CG or block norm-based lower bounding procedures) to re-update multipliers and obtain a valid lower bound on z *. Note that, this procedure reduces to the classical SL algorithm when z H (β, μ)>z UB best holds at each step. Otherwise, z H (β, μ) mimics as if it is a true lower bound and the algorithm converges as usual. However, according to our computational experiments we have not observed such a case where z H (β, μ) has exceeded z UB best. At the end of the SL algorithm, we have performed one of the recovery procedures (ie CG with DC, CG with WPLD, ℓ 1 UDAP and ℓ ∞ UDAP) in order to ensure that the algorithm produces a valid lower bound on z *.
A.2. Multi-commodity-capacitated alternate location allocation heuristic
Pursuing the basic ideas proposed by Cooper (1972), the MCALA heuristic is described below. Observe that once a feasible transportation plan is given, MCMWP reduces to solving I Weber problems
for each facility i=1, …, I. Here . Note that each of these I Weber problems can be solved by Weiszfeld's algorithm (1937) and its generalizations. Although the summation is taken over all customers, it considers only I i of them, which is the size of the set defined by I i =∣{(j, k) : w ijk >0}∣. Clearly, ∑ i=1 I I i ⩾J × K holds since a customer can be served by more than one facility. In short, when a feasible assignment of w ijk variables is given, the problem reduces to optimally locating single facilities with respect to I i customers. Consequently, it is possible to tailor Cooper's (1964) ALA heuristic to produce a good feasible solution for MCMWP. We give below a formal outline of the MCALA heuristic:
Algorithm 2 MCALA Heuristic
-
1
Locate I facilities at randomly selected points x i =( x i 1 x i 2 ) T for i =1, …, I ;
-
2
For each facility i and customer j calculate the distance d ( x i , a j ) between them, and set the new unit transportation costs as c̄ ijk = c ijk d ( x i , a j );
-
3
Determine feasible allocations w ijk * by solving the Multi-commodity Transportation Problem (MTP):
-
4
Solve I Weber problems (A.1) to relocate I facilities;
-
5
Repeat Steps 3 and 4 until either facility locations x i =(x i1 x i2 ) T for i=1, …, I or allocations w ijk * for i=1, …, I, j=1, …, J, and k=1, …, K remain unchanged.
In order to observe and compare the performance of our LR strategies and the lower bounds they produce, we have adapted the Sherali et al's (2002) RLT-based lower bounding approach to the multi-commodity extension of the CMWP, that is, the MCMWP. The resulting RLT formulation is presented in the following.
A.3. Derivation of the lower bounding RLT formulation
The MCMWP formulation given by (1)–(5) can be rewritten as
where the allocation variables have lower and upper bounds which can be set as l ijk =0 and u ijk =min {s ik , q jk , u ij } for i=1, …, I; j=1, …, J; k=1, …, K and the new defined variables α ij measure the distance between facility i and customer j. The RLT approach is applied in order to get a lower bounding linear programming formulation on the formulation given by (A.2)–(A.7), which has bilinear objective function and non-linear constraints. The non-linear constraint set is approximated by generating some supports on it.
Let the facets of the convex hull of the customer locations be defined by a set of inequalities of the form
Let us define the following bounds on α ij :
The inequalities given above are linearized which result in the following set of constraints:
Reformulation:
-
1
Multiply each inequality given by (A.12) with each of inequality (A.7).
-
2
Multiply each inequality given by (A.8), (A.9), (A.10) by each of inequality (A.7).
-
3
Generate the following equalities:
Linearization:
The resulting constraints are linearized by substituting the bilinear terms which are defined with the following new decision variables:
Note that the notation used by Sherali et al (2002), where additional details can be found within the context of the CMWP, is maintained here. We present next the RLT-based lower bounding formulation for the MCMWP:
RLT:
Rights and permissions
About this article
Cite this article
Akyüz, M., Öncan, T. & Altınel, İ. Solving the multi-commodity capacitated multi-facility Weber problem using Lagrangean relaxation and a subgradient-like algorithm. J Oper Res Soc 63, 771–789 (2012). https://doi.org/10.1057/jors.2011.81
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1057/jors.2011.81