An improved statistical approach for consistency test in AHP

The pair-wise comparison matrix (PCM) is widely used in multi-criteria decision making methods. If the PCM is inconsistent, the resulting priority vector is not reliable. Hence, it is necessary to measure the level of the inconsistency of the PCM. There are two approaches for testing the consistency of the PCM: deterministic approaches and statistical or stochastic approaches. In this paper, an improved statistical approach to test the consistency of the PCM is proposed, which combines hypothesis test and maximum likelihood estimation. The proposed statistical approach is flexible and reliable because it sets a suitable significance level according to different situations. Two numerical examples are introduced to illustrate the proposed statistical approach.     

Enhancing data consistency in decision matrix: Adapting Hadamard model to mitigate judgment contradiction

Cardinal and ordinal inconsistencies are important and popular research topics in the study of decision making with pair-wise comparison matrices (PCMs). Few of the currently-employed tactics are capable of simultaneously dealing with both cardinal and ordinal inconsistency issues in one model, and most are heavily dependent on the method chosen for weight (priorities) derivation or the obtained closest matrix by optimization method that may change many of the original values. In this paper, we propose a Hadamard product induced bias matrix model, which only requires the use of the data in the original matrix to identify and adjust the cardinally inconsistent element(s) in a PCM. Through graph theory and numerical examples, we show that the adapted Hadamard model is effective in identifying and eliminating the ordinal inconsistencies. Also, for the most inconsistent element identified in the matrix, we develop innovative methods to improve the consistency of a PCM. The proposed model is only dependent on the original matrix, is independent of the methods chosen to derive the priority vectors, and preserves most of the original information in matrix A since only the most inconsistent element(s) need(s) to be modified. Our method is much easier to implement than any of the existing models, and the values it recommends for replacement outperform those derived from the literature. It significantly enhances matrix consistency and improves the reliability of PCM decision making.     

A simple formula to find the closest consistent matrix to a reciprocal matrix

Achieving consistency in pair-wise comparisons between decision elements given by experts or stakeholders is of paramount importance in decision-making based on the AHP methodology. Several alternatives to improve consistency have been proposed in the literature. The linearization method (Benítez et al., 2011 [10]), derives a consistent matrix based on an original matrix of comparisons through a suitable orthogonal projection expressed in terms of a Fourier-like expansion. We propose a formula that provides in a very simple manner the consistent matrix closest to a reciprocal (inconsistent) matrix. In addition, this formula is computationally efficient since it only uses sums to perform the calculations. A corollary of the main result shows that the normalized vector of the vector, whose components are the geometric means of the rows of a comparison matrix, gives the priority vector only for consistent matrices.     

A heuristic approach for deriving the priority vector in AHP

This paper proposes an approach for deriving the priority vector from an inconsistent pair-wise comparison matrix through the nearest consistent matrix and experts judgments, which enables balancing the consistency and experts judgments. The developed algorithm for achieving a nearest consistent matrix is based on a logarithmic transformation of the pair-wise comparison matrix, and follows an iterative feedback process that identifies an acceptable level of consistency while complying with experts preferences. Three numerical examples are examined to illustrate applications and advantages of the developed approach.     

A Goal Programming Method for Generating Priority Vectors

The generation of priority vectors from pairwise comparison information is an integral part of the Analytic Hierarchy Process (AHP). Traditionally, either the right eigenvector method or the logarithmic least squares method have been used. In this paper, a goal programming method (GPM) is presented that has, as its objective, the generation of the priority vector whose associated comparison values are, on average, the closest to the pairwise comparison information provided by the evaluator. The GPM possesses the properties of correctness in the consistent case, comparison order invariance, smoothness, and power invariance. Unlike other methods, it also possesses the additional property that the presence of a single outlier cannot prevent the identification of the correct priority vector. The GPM also has a pair of naturally meaningful consistency indicators that offer the opportunity for empowering the decision maker. The GPM is thus an attractive alternative to other proposed methods.     

A simple method to improve the consistency ratio of the pair-wise comparison matrix in ANP

Tests of consistency for the pair-wise comparison matrices have been studied extensively since AHP was introduced by Saaty in 1970s. However, existing methods are either too complicated to be applied in the revising process of the inconsistent comparison matrix or are difficult to preserve most of the original comparison information due to the use of a new pair-wise comparison matrix. Those methods might work for AHP but not for ANP as the comparison matrix of ANP needs to be strictly consistent. To improve the consistency ratio, this paper proposes a simple method, which combines the theorem of matrix multiplication, vectors dot product, and the definition of consistent pair-wise comparison matrix, to identify the inconsistent elements. The correctness of the proposed method is proved mathematically. The experimental studies have also shown that the proposed method is accurate and efficient in decision maker’s revising process to satisfy the consistency requirements of AHP/ANP.     

Fuzzy analytic hierarchy process: A logarithmic fuzzy preference programming methodology

Fuzzy analytic hierarchy process (AHP) proves to be a very useful methodology for multiple criteria decision-making in fuzzy environments, which has found substantial applications in recent years. The vast majority of the applications use a crisp point estimate method such as the extent analysis or the fuzzy preference programming (FPP) based nonlinear method for fuzzy AHP priority derivation. The extent analysis has been revealed to be invalid and the weights derived by this method do not represent the relative importance of decision criteria or alternatives. The FPP-based nonlinear priority method also turns out to be subject to significant drawbacks, one of which is that it may produce multiple, even conflict priority vectors for a fuzzy pairwise comparison matrix, leading to entirely different conclusions. To address these drawbacks and provide a valid yet practical priority method for fuzzy AHP, this paper proposes a logarithmic fuzzy preference programming (LFPP) based methodology for fuzzy AHP priority derivation, which formulates the priorities of a fuzzy pairwise comparison matrix as a logarithmic nonlinear programming and derives crisp priorities from fuzzy pairwise comparison matrices. Numerical examples are tested to show the advantages of the proposed methodology and its potential applications in fuzzy AHP decision-making.     

Fuzzy multi-criteria decision-making based on positive and negative extreme solutions

To encompass decision data vagueness, many researchers generalized multi-criteria decision-making (MCDM) methods in certain environment into fuzzy multi-criteria decision-making (FMCDM) methods under fuzzy environment. In these FMCDM methods, ranking fuzzy numbers based on fuzzy pair-wise comparison is normally essential, but the comparison is a complexity work. To avoid fuzzy pair-wise comparison, we propose a FMCDM method based on positive and negative extreme solutions of alternatives. In the proposed method, two extreme solutions of alternatives are obtained by MAX and MIN operations of fuzzy TOPSIS. Then weakness and strength matrices between alternatives and extreme solutions are derived by a difference function revised from fuzzy preference relation of Lee, and multiplied with weight matrix to be weighted weakness and strength indices. The two weighted indices are respectively transferred into positive and negative indices, and then the two indices integrated into a total performance index. Finally, alternatives can be sorted according to their related performance indices, and FMCDM problems are easily solved, not by fuzzy pair-wise comparison.     

Maximum likelihood least squares identification for systems with autoregressive moving average noise

Maximum likelihood methods are important for system modeling and parameter estimation. This paper derives a recursive maximum likelihood least squares identification algorithm for systems with autoregressive moving average noises, based on the maximum likelihood principle. In this derivation, we prove that the maximum of the likelihood function is equivalent to minimizing the least squares cost function. The proposed algorithm is different from the corresponding generalized extended least squares algorithm. The simulation test shows that the proposed algorithm has a higher estimation accuracy than the recursive generalized extended least squares algorithm.     

基于基本回路修正的AHP一致性调整方法研究

为解决AHP一致性问题,提出一种基于基本回路修正的调整方法,能够同时解决数值不一致和逻辑不一致问题,同时保证对原始信息的修改量最小。数值不一致和逻辑不一致均由决策者的不准确判断引起,其中数值不一致可以通过降低一致性比率(CR)值进行改善,而逻辑不一致只有将判断矩阵中所有三阶回路去除才能得到解决。因此,通过对n阶判断矩阵进行基本矩阵分解,得到C³_n个3阶的基本矩阵,其中存在三阶回路的称为基本回路,从而将判断矩阵的一致性修正问题转化为基本回路的一致性修正问题。通过对基本回路的一致性比较,提出了两种确定最不一致元素的方法,即CR和最大法和优化法,并设计了优化模型对最不一致元素进行修正。最后,通过算例分析验证了本文方法的可行性,与已有方法的对比结论证明了本文方法更为有效。