Instances for the Capacitated Mobile Facility Location Problem

These data files correspond to p-med instances in Raghavan, S., Sahin, M., and Salman, F. S., "The Capacitated Mobile Facility Location Problem", published in the European Journal of Operational Research (https://doi.org/10.1016/j.ejor.2019.02.055).

The zip file contains two folders named "heterogeneous_facilities" and "homogeneous_facilities". Each of these folders contains further subfolders named "large_C_over_F" (where the |C|/|F| ratio is greater than 10) and "small_C_over_F" (where the |C|/|F| ratio is less than or equal to 10).

There are 28 instances in the "small_C_over_F folder" and 17 instances in the "large_C_over_F" folder. The tables in the paper report results for each of 28 instances (both homogeneous and heterogenous) in the "small_C_over_F" folders.

The format of these data files is:

|V|

d_{1,1} d_{1,2} ... d_{1,|V|}
d_{2,1} d_{2,2} ... d_{2,|V|}
...
d_{|V|,1} d_{|V|,2} ... d_{|V|,|V|}

w_{1} w_{2} ... w_{|V|}

Q_{1} Q_{2} ... Q_{|V|}

u_{1} u_{2} ... u_{|V|}

q_{1} q_{2} ... q_{|V|}

where:

|V| is the number number of vertices;
d_{i,j} (i,j = 1,...,|V|) is the distance between i and j
w_{j} (j = 1,...,|V|) is the facility locations and weights, if w_{j} > 0, there is a facility located at j and w_{j} is the facility weight
Q_{j} (j = 1,...,|V|) is the facility locations and capacities, if Q_{j} > 0, there is a facility located at j and Q_{j} is the facility capacity
u_{j} (j = 1,...,|V|) is the client locations and weights, if u_{j} > 0, there is a client located at j and u_{j} is the client weight
q_{j} (j = 1,...,|V|) is the client locations and demands, if q_{j} > 0, there is a client located at j and u_{j} is the client demand

Further details on these instances are provided in the paper mentioned above.

本数据集文件对应拉格万（Raghavan, S.）、萨欣（Sahin, M.）与萨尔曼（Salman, F. S.）发表于《欧洲运筹学杂志》（European Journal of Operational Research，DOI: 10.1016/j.ejor.2019.02.055）的论文《容量约束移动设施选址问题》（The Capacitated Mobile Facility Location Problem）中的p-med实例（p-med）。

该压缩包包含两个文件夹，分别命名为"heterogeneous_facilities（异构设施）"和"homogeneous_facilities（同构设施）"。每个文件夹下均设有子文件夹"large_C_over_F"（其中|C|/|F|比值大于10）与"small_C_over_F"（其中|C|/|F|比值小于或等于10）。

"small_C_over_F"文件夹内包含28个实例，"large_C_over_F"文件夹内包含17个实例。该论文中的表格报告了"small_C_over_F"文件夹下28个实例（涵盖同构与异构设施场景）的求解结果。

本数据集文件的格式如下：
|V|
d_{1,1} d_{1,2} … d_{1,|V|}
d_{2,1} d_{2,2} … d_{2,|V|}
……
d_{|V|,1} d_{|V|,2} … d_{|V|,|V|}
w_{1} w_{2} … w_{|V|}
Q_{1} Q_{2} … Q_{|V|}
u_{1} u_{2} … u_{|V|}
q_{1} q_{2} … q_{|V|}

其中，|V|为顶点数量；d_{i,j}（i,j = 1,…,|V|）为顶点i与顶点j之间的距离；w_{j}（j = 1,…,|V|）代表设施位置与权重：若w_j > 0，则顶点j处设有设施，w_j即为该设施的权重；Q_{j}（j = 1,…,|V|）代表设施位置与容量：若Q_j > 0，则顶点j处设有设施，Q_j即为该设施的容量；u_{j}（j = 1,…,|V|）代表客户位置与权重：若u_j > 0，则顶点j处设有客户，u_j即为该客户的权重；q_{j}（j = 1,…,|V|）代表客户位置与需求：若q_j > 0，则顶点j处设有客户，q_j即为该客户的需求。

有关该类实例的更多详细信息，请参见前文提及的学术论文。