Doi: 10. 1016/S0305-0548(03)00202-8
Download 490.36 Kb. Pdf ko'rish
|
10.1016@S0305-05480300202-8
1.26
1.36 1.66 1.44 1.67 1.43 1.88 1.68 1.76 1.21 1.74 1.59 1.29 1.41 1.1 1.55 1.8 1.53 1.1 6.42 1.66 1.59 1.26 2.65 t = 10.33 1.43 1.36 1.29 4.06 3.12 6.17 1.68 4.64 2.88 1.67 t = 13.35 Fig. 1. Sample network with two aggregation trees. The numbers stand for the tra3c of the links or nodes. The picture on the right shows an access network, which can be built based on the input given on the left. that the addition operator is de/ned for it. The maximum number of levels (depth constraint) of the resulting aggregation trees is given by L ¿ 2; the roots are in level 1. The maximum indegree of the nodes in each level is given by D k ¿ 0; k =1; : : : ; L, where D L =0 as the lowest level leaf nodes have no incoming links. The input function C Eq (i; l; t) returns the cost of the equipment to be installed into node i ∈ N, if it is placed at level 1 6 l 6 L in an aggregation tree, and if it aggregates t ∈ T amount of tra3c. The input function C Li (i; j; t) returns the cost of a link between nodes i; j ∈ N, i = j, if it carries t ∈ T aggregated tra3c. 2.3. Output The task is to interconnect the set of nodes to form a set of trees with the least possible cost. The decision variable x ij ∈ {0; 1} equals to 1 i? node i is connected upwards to node j in the aggregation tree containing it, ∀i; j ∈ N. Variable 1 6 l i 6 L denotes the level of node i in the tree containing it, where l i = 1 for a root node, l i = 2 for a child node of a root, etc., ∀i ∈ N. We derive some auxiliary variables from these two sets of variables: l ik ∈ {0; 1} equals to 1 i? node i is at level k in the tree containing it, ∀i ∈ N, k = 1; : : : ; L, while t i ∈ T denotes the total aggregated tra3c in node i, ∀i ∈ N. 2.4. Problem formulation The access network planning problem is then: min n i=1 n j=1;j =i x ij C Li (i; j; t i ) + n i=1 C Eq (i; l i ; t i ) (1) with respect to l ik = 1 if l i = k; 0 if l i = k; ∀i ∈ N; k ∈ {1; : : : ; L}; (2) t j = T j + n i=1;i =j x ij t i ∀j ∈ N; (3) 62 I. G-odor, G. Magyar / Computers & Operations Research 32 (2005) 59–86 n j=1;j =i x ij = 1 − l i1 ∀i ∈ N; (4) n i=1;i =j x ij 6 L k=1 l jk D k ∀j ∈ N; (5) x ij = 1 ⇒ l i = l j + 1 ∀i; j ∈ N; (6) x ij ∈ {0; 1} ∀i; j ∈ N; i = j; (7) l i ∈ {1; : : : ; L} ∀i ∈ N; (8) l ik Download 490.36 Kb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling