Discrete Dynamics in Nature and Society

Volume 2018, Article ID 1610349, 8 pages

https://doi.org/10.1155/2018/1610349

## Geometric Programming with Discrete Variables Subject to Max-Product Fuzzy Relation Constraints

^{1}School of Mathematics and Information Science, Guangzhou University, Guangzhou, Guangdong 510006, China^{2}Guangzhou Vocational College of Science and Technology, Guangzhou University, Guangzhou, Guangdong 510550, China^{3}Department of Industrial and System Engineering, North Carolina State University, Raleigh, NC 27695, USA^{4}Department of Mathematics and Statistics, Hanshan Normal University, Chaozhou, Guangdong 521041, China

Correspondence should be addressed to Bingyuan Cao; moc.361@ygniboac

Received 18 December 2017; Accepted 13 February 2018; Published 12 April 2018

Academic Editor: Seenith Sivasundaram

Copyright © 2018 Zejian Qin et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

#### Abstract

The problem of geometric programming subject to max-product fuzzy relation constraints with discrete variables is studied. The major difficulty in solving this problem comes from nonconvexity caused by these product terms in the general geometric function and the max-product relation constraints. We proposed a 0-1 mixed integer linear programming model and adopted the branch-and-bound scheme to solve the problem. Numerical experiments confirm that the proposed solution method is effective.

#### 1. Introduction

Since fuzzy relation equations with max-min composition were firstly introduced by Sanchez [1–3], they have attracted much research attention. As an extension, fuzzy relation inequalities associated with max--norm were also studied. As demonstrated in [4], the complete solution set of max--norm fuzzy relation equations can be completely determined by a unique maximum solution and a finite number of minimal solutions. It is easy to compute the maximum solution, but finding all the minimal solutions is an NP-hard problem [4–6]. It is worth mentioning that Li and Fang [5] provided a complete survey and detailed discussion on fuzzy relational equations. They studied fuzzy relational equations in a general lattice-theoretic framework and introduced classification of basic fuzzy relational equations.

Meanwhile, optimization problems subject to fuzzy relation equations or inequalities were introduced and studied. Fang and Li [7] firstly investigated a linear optimization problem with a consistent system of max-min equations. They converted it into a 0-1 integer programming problem and solved this by the branch-and-bound method. Then it became a unified framework to deal with the resolution of linear optimization problem subject to a system of fuzzy relational equations with max--norm composition. And linear programming problem with fuzzy relation constraints became a hot topic for further research [8–10]. Guo and Xia [11] proposed a method to accelerate the resolution of this problem. Zhang et al. [12] studied a linear objective optimization problem with max-min fuzzy relation inequalities.

For nonlinear programming with fuzzy relation constraints, many achievements were gained. Wang et al. [13] first studied latticized linear programming subject to max-min fuzzy relation inequalities. In [14] Li and Fang made a further study on the latticized linear optimization (LLO) problem and its variant, which are a special class of optimization problems constrained by fuzzy relational equations or inequalities. Yang and Cao [15] and Wu [16] considered geometric optimization problems with single-term exponents under fuzzy relation equation constraints with max-min composition, where the objective function is . Zhou and Ahat [17] investigated similar problem where the composition was replaced by max-product. Yang et al. [18] investigated min-max programming problem subject to fuzzy relation inequality constraints with a special composition of addition-min. Yang et al. [19] studied the single-variable term semi-latticized geometric programming subject to max-product fuzzy relation equations. As important nonlinear programming, geometric programming with fuzzy relational constraints attracted some researchers’ attention. Yang and Cao [20] proposed a monomial geometric programming subject to max-min fuzzy relation equations with the objective function being . Shivanian and Khorram [21] considered monomial geometric programming subject to fuzzy relation inequalities with max-product composition. Zhou et al. [22] investigated a special posynomial geometric programming problem subject to max-min fuzzy relation equations, in which the exponents of each variable are all nonpositive or nonnegative real numbers. Aliannezhadi et al. [23] investigated a monomial geometric programming objective function subject to bipolar max-product fuzzy relation constraints. All the objective functions of these optimization problems are special geometric functions. Because of the nonconvexity and complexity, problems with a general geometric objective function have not been studied. At the same time, the data in the real world could often be discrete instead of continuous. In particular, statistics data are often discrete numbers associated with real facts. Therefore, in this paper we study geometric programming subject to max-product fuzzy relational constraints, in which the objective function is a general geometric function and all the variables are of discrete values.

The rest of the paper is organized as follows. In Section 2, we introduce some basic concepts. A method to transform the original problem into a linear mix-integer programming model is proposed in Section 3. Numerical experiments are given to illustrate the effectiveness of the proposed solution method in Section 4. A simple conclusion is arranged in Section 5.

#### 2. Problem

Consider the following geometric programming problem subject to max-product fuzzy relational constraints with discrete variables:where , and are three index sets, and , for . The notation “” denotes the maximum operator and “” can be “≤”, “≥”, or “=” for upper-bound constraints, lower-bound constraints, and equation constraints, respectively.

We analyze the three kinds of max-product fuzzy relation constraints with discrete variables first.

(i) For an upper-bound constraint:It can be written in constraints:Hence , where the notation “” denotes the minimum operator. Combining with the discrete requirements: we have By checking and , we can reset the discrete values of each variable.

(ii) For an equation constraint:It can be written as an upper-bound constraint and a lower-bound constraint, that is, From (i) we know that the upper-bound constraint can be included in a new set of discrete values, and the lower-bound constraint can be grouped into other lower-bound constraints. Consequently, the constraints can all be transformed into max-product lower-bound constraints with discrete variables. Without loss of generality, in this paper we consider the following geometric programming problem subject to max-product fuzzy relational inequalities with discrete variables:where , for .

The major difficulty in solving problem (8) comes from the nonconvexity caused by the cross-product terms in the objective function and the max operation in the fuzzy inequalities. We will introduce methods to transform the original problem into an equivalent 0-1 linear mixed integer programming problem and adopt the branch-and-bound scheme to find an optimal solution.

#### 3. Linear Reformulation

##### 3.1. Treating Geometric Function with Multivalued Discrete Variables

For solving problem (8), the first step is to convert each multivalued variable into 0-1 valued by consideringwhere is an index set and for and .

It is important to know that the binary variables of and , induced by (9), may cause heavy computational burden when and become large. To avoid this burden, Li et al. [24] proposed a logarithmic approach which needs only binary variables ( denotes the ceiling function), nonnegative variables, and linear equations to represent the discrete variables with values by consideringwhere is an index set; is a binary variable; is now a nonnegative variable, for and ; and are binary variables obtained by solving the equations of for .

The next step is to reformulate the objective function. Remembering that and , we denote for ; then we haveMoreover, we denote Then the objective function can be expressed asIn this way we can express the objective function of problem (8) as a linear function in . Note that the expression (12) is nonlinear, but by adopting the approach in [25], we can express them by using linear inequalities in in the next result.

Lemma 1. *Given , the product term , with , can be expressed by the following linear inequalities in and :where is defined in (9), and is a sufficient large positive value such that*

*Proof. *Assume that for a particular . From (9), we have and other .

(i) For , inequality (14) can be converted into the following equation: (ii) For , inequality (14) can transformed asActually, from (i), we have From (15), we know and ; therefore, Hence, From the above, we can see, when , (14) will be converted into (12); when , (14) will be converted into the redundant constraints (17). Consequently (12) can be expressed by (14).

##### 3.2. Treating Max-Product Fuzzy Relation Inequalities

Firstly we recall some basic concepts and important properties of the max-product fuzzy inequalities: Denote and ; “” represent the max-product composite operation; then a system of max-product fuzzy relation inequalities can be represented in the following matrix form:The set of all the solutions of system (22) is called its solution set, denoted by . If , we say the system (22) is consistent; otherwise it is inconsistent [4, 26].

Lemma 2 (see [5, 26, 27]). *Let be a system of consistent max-product fuzzy relation inequalities; if and , then we know .*

*Definition 3 (see [26]). *A solution is said to be the maximum or greatest solution if and only if for all . A solution is said to be a minimal solution if and only if implies for any .

Obviously, if the maximum solution of (22) exists, it is unique and .

Theorem 4 (see [26, 27]). *Let be a system of max-product fuzzy relation inequalities; then it is consistent if and only if . Moreover, if the system is consistent, the solution set can be fully determined by one maximum solution and a finite number of minimal solutions, that is, where is the set of all minimal solutions of (22).*

The feasible domain restricted by the max-product fuzzy relation inequalities (22) can be fully characterised by the solution set . Actually, if we know all the minimal solutions, then the solution set can be expressed. The main problem we face is computing all the minimal solutions of system (22). However this is an NP-hard problem [5, 26]. Hence we introduce a method to find all the potential minimal solutions instead of minimal solutions.

*Definition 5 (see [26]). *Matrix is called a discrimination matrix of (22) with

Theorem 6 (see [26]). *Let matrix be the discrimination matrix of the system (22). The system is consistent if and only if, for any , there exists at least one such that .*

*Definition 7 (see [27]). *Let be the discrimination matrix of the system (22) and matrix , where . We call a solution matrix if, for any , there exists a unique such that .

Let be a solution matrix of the system (22); denote , where

Theorem 8 (see [27]). *Let be a solution matrix of the system (22) and be defined by (25); then . Moreover, for any , there exists a solution matrix with a corresponding such that .*

From Theorem 8, we know that, for any minimal solution of (22), the solution matrix and its corresponding can be formed such that . Then the feasible domain restricted by the max-product fuzzy relation inequalities constraints (22) can be expressed aswhere denotes the set of all the solution matrix determined by discrimination matrix of system (22), and is a solution corresponding to the solution matrix .

Consider the definition of (25); we see that, for any solution matrix and its corresponding , the inequality is equivalent to the following constraints:At the same time, we bring in binary variables , for and ; the element of matrix can be expressed aswhere and

Therefore, the feasible domain restricted by (22) can be characterised bywhere and

Adopting the method introduced above, we have a linear reformulation of the geometric objective function and max-product fuzzy relation inequality constraints. The original problem (8) can then be converted into the following 0-1 linear mixed integer programming model: Note that, in this model, there are binary variables of ; binary variables of ; nonnegative variables of ; nonnegative variables of ; nonnegative variables of ; linear equations (i.e., (9) + (10) + (11) + (29)); and linear inequality constraints (i.e., (14) + (30)) involved.

Now we illustrate this model by a simple example.

*Example 9. *Consider the following geometric programming problem in discrete variables with max-product fuzzy relation inequality constraints:In this case, , adopting the logarithmic approach (10) and solving the equations of for ; we haveUsing (9) and (10), we haveBy (24) and (30), we have Then by applying (29) and (30), we haveFinally, adding (11) and (14), problem (31) can be converted into the following model:Solving problem (36) leads to a solution with , , and for other variables. The corresponding optimal objective value is −2.4813.

#### 4. Computational Experiments

To illustrate the effectiveness of the proposed method, we conduct some computational experiments. All the experiments have been run on a PC equipped with the Intel Core I7-6700HQ CPU, 8 GB RAM, and Windows 10 (64 bit) operating system. GUROBI (2016) is the chosen MIP solver for solving all the instances. To ensure the existence of an optimal solution in each case, we apply the following rules to randomly generate its coefficients:

are generated randomly by the uniform distribution over , and are taken randomly from for and .

are generated randomly by the uniform distribution over such that for and .

When , the elements and may go to infinite; in order to avoid this, we generate randomly in the interval such that for and .

To make sure the test instance is feasible, each is generated randomly by the uniform distribution over the interval of for . Our computational experiments consist of cases with and . Each case includes 20 test instances which are executed with the coefficients generated by the rules mentioned above.

The computational results are shown in Table 1. In the Table, V-Bin denotes the number of 0-1 variables, V-Con denotes the number of continuous variables, L-Eq denotes the number of linear equation constraints, L-Ine denotes the number of linear inequality constraints, denotes the average CPU time in seconds of solving 20 randomly generated instances of each case, and means the longest CPU time in seconds we meet in each case.