Application of fuzzy calculations for improving portfolio matrices
Introduction
Strategic management is defined as the science of formulating, executing, and assessing those decisions that enable organizations to achieve their long-term goals [4]. The most important stage in this process deals with strategy formulation. In this stage, the organization’s mission is determined, and threats and opportunities are identified. Strengths and weaknesses are identified, long-term goals are determined, various strategies are proposed, and specific strategies are selected.
One of the most important instruments used in this stage is the business portfolio analysis matrix [3], [4], [8], [13]. The matrix is composed of zones, which represent predefined sets of strategies. The evaluation of the firm’s internal and external conditions results in a position (usually a point) in one of the strategy zones. The industry attractiveness-business strength matrix is the variation of the business portfolio analysis matrix that we have selected for the purposes of this research (see Fig. 1) [4].
The strategies represented by each zone are as follows:
- (1)
Cells 1, 2 and 4 (zone A) comprise strategies of growth and building, and also offensive strategies. These strategies are known as “intensive strategies,” i.e., market penetration, market development, product development, forward integration, backward integration and horizontal integration.
- (2)
Cells 3, 5 or 7 (zone B) call for the preservation of the status quo. In this zone, common strategies are market penetration and product development.
- (3)
Units located in cells 6, 8 or 9 (zone C) should divest.
In the above matrix, point calculations are usually based on inexact terms, but fuzzy-based methodologies have seldom been used. The application of fuzzy theory to the interpretation of portfolio matrices is recommended in [11], [12]. In [11], the industry attractiveness-business strength matrix is fuzzified. In [12] a fuzzy rule-based method is used to handle the growth-share matrix. The fuzzy concept also has been applied in optimal portfolio selection problems. For instance, in [9], the portfolio selection problem is investigated when the security returns contain both randomness and fuzziness. In [7], a fuzzy mathematical model is applied to a similar problem. In [1], a fuzzy random multi-objective quadratic programming method is examined and applied to the portfolio optimization problem. Finally, in [14], the analytic hierarchy process is applied to the portfolio selection problem.
Despite this extensive research, difficulties still persist: portfolio analysis recommends a strategy for each business unit based on its position in the company’s overall portfolio of businesses, according to known and accepted rules. This could lead to different strategy recommendations for business units that are very close to each other but on the opposite sides of boundaries in matrices. This last point can be construed as the first drawback of portfolio analysis. Also, the portfolio analysis suggests the same strategy choice for all business units in the same quadrant, regardless of their exact positions in the matrix. For example, one of the legitimate shortcomings of the growth-share matrix is that a “four-cell matrix based on high/low classification hides the fact that many businesses are in markets with an average growth rate and have relative market shares that are neither high nor low but in-between or intermediate. In which cells do these average businesses belong?” [12].
The expression of internal and external factors in the form of fuzzy numbers, instead of crisp numbers, provides a convenient answer to that question [5]. Therefore, one should consider an environment for the expression of strategies based on areas instead of points. Although some researchers have attempted to fuzzify the inputs of process of strategic planning [11], [12], this paper considers both inputs and outputs of this process. Whereas the real nature of the environment for the implementation of strategies is ambiguous, we pursue an approach towards the expression of strategies in uncertain states. The ability to prioritize strategies is one of the outcomes of such an approach. Furthermore, by quantifying the factors as fuzzy triangular numbers, the evaluation of the factors is made more flexible, and both qualitative and quantitative aspects of the factors can be considered. The ambiguity and uncertainty of the factors can also be taken into account during the process of decision making. Therefore, our aim in this paper is to develop a flexible, adaptive, and realistic system to support the process of decision making.
Section snippets
Fuzzy sets application in industry attractiveness-business strength matrix
There are three main problems with the classical method of strategy extraction using the industry attractiveness-business strength matrix:
- (1)
The internal factor evaluation (IFE) matrix is used to evaluate the internal conditions of an organization. The IFE matrix comprises factors (strengths and weaknesses), weights (0.0–1.0), ratings (1–4: One for high weakness, two for low weakness, three for low strength, and four for high strength), and the final weighted score is obtained by multiplying the
Factors evaluation
Suppose that there are n SBUs to be evaluated considering m internal and k external factors. Suppose that Wi is the fuzzy weight number representing the importance of the internal factor Fi and Dij is the desirability of the jth SBU regarding that factor, for all and for all . The overall internal desirability of the SBU j is denoted by IDj.
The -cut of the overall internal desirability IDj of a SBUj is calculated through fuzzy weight average (FWA) [15] for m
A case illustration
In order to evaluate the applicability of the proposed algorithm, we implemented it in a strategic planning process for a food company in Iran [19]. The company is a successful player in the food industry in the Middle East.
Conclusion
Portfolio analysis using a fuzzy approach has been the main focus of this paper. We have drawn on fuzzy methodology to extract strategies. Fuzzy methodology allows us to incorporate uncertainty into historical data and also to incorporate subjective/intuitive characteristics into the portfolio analysis models. Furthermore, the ambiguity and uncertainty of the factors can be taken into account during the process of decision making.
The industry attractiveness-business strength matrix is a
References (19)
On solutions of fuzzy random multi-objective quadratic programming with applications in portfolio problem
Information Sciences
(2008)- et al.
Asset portfolio optimization using fuzzy mathematical programming
Information Sciences
(2008) A new perspective for optimal portfolio selection with random fuzzy returns
Information Sciences
(2007)- et al.
Application of fuzzy sets with different t-norms in the interpretation of portfolio matrices in strategic management
Fuzzy Sets and Systems
(2000) - et al.
Fuzzy portfolio selection using fuzzy analytic hierarchy process
Information Sciences
(2009) Is there a need for fuzzy logic
Information Sciences
(2008)Toward a generalized theory of uncertainty (GTU) – an outline
Information Sciences
(2005)- et al.
National program for cleaner production (CP) in Iran: a framework and draft
Journal of Cleaner Production
(2006) - et al.
A fuzzy logic based decision making approach for new product development
International Journal of Production Economics
(2002)
Cited by (25)
A new approach to the bi-criteria multi-period fuzzy portfolio selection
2021, Knowledge-Based SystemsCitation Excerpt :As the results, a rapid growth of interest to the use of fuzzy sets, the theory of possibility with their different generalizations and modifications for the formulation and solution of PS problems has being noted during the last two decades. The classical fuzzy set theory was used for PS in [16–19]. The intuitionistic [20] and hesitant [21,22] fuzzy models were applied to solve PS problems.
Portfolio model for analyzing human resources: An approach based on neuro-fuzzy modeling and the simulated annealing algorithm
2017, Expert Systems with ApplicationsCitation Excerpt :However, the BCG matrix suggests the same strategy of selection for all the elements in the same matrix quadrant, not considering the exact position of the elements in the matrix. In order to overcome these difficulties, the authors Ghazinoory et al. (2010), Oh, Yang, and Lee (2012) and Kocadagli and Keskin (2015) suggest the application of fuzzy logic. The aim of this paper is to overcome the aforementioned gap using adaptive neuro-fuzzy techniques and the simulated annealing algorithm for creating the human resources portfolio.
Cost and risk aggregation in multi-objective route planning for hazardous materials transportation—A neuro-fuzzy and artificial bee colony approach
2016, Expert Systems with ApplicationsPersonalized finance advisory through case-based recommender systems and diversification strategies
2015, Decision Support SystemsSome new results on value ranges of risks for mean-variance portfolio models
2013, Information SciencesPB-ADVISOR: A private banking multi-investment portfolio advisor
2012, Information SciencesCitation Excerpt :In this area, the use of genetic algorithms (e.g. [13,41,42,52]), fuzzy logic (e.g. [31,36,53,54,59,63,73,77]), neural networks (e.g. [20,76,25,47]), decision support systems (e.g. [40,53,37,65]) among several other techniques has proven to be just as effective in financial portfolio selection. Focusing on fuzzy logic, one of the components of PB-ADVISOR, several authors have recommended the incorporation of fuzzy theory into the interpretation of portfolio matrices with different applications: e.g. to optimize the selection of portfolio products [31], to obtain optimal proportions of the assets for constructing a portfolio that respects investor-preferences [36], for strategic management [53], to assist managers in reaching a better understanding of the overall competitiveness of their business portfolios [54], to facilitate the interpretation of portfolio matrices used for strategy alternative(s) formulation and selection [59], to solve the portfolio problem when security returns are bifuzzy variables [73] and to define new dynamic portfolio selection strategy in the stock market [77]. Given that the investments market is a very complex environment, returns cannot always be accurately predicted from historical data, as they are beset with ambiguity and vagueness [41].