Abstract
Any company must constantly innovate if they want to maintain its market share in the present cutthroat and unstable industry. Innovation has a big influence on consumer behavior, yet it goes against the principles of sustainability. The issue of sustainability has become crucial to their company’s growth. In order to evaluate a business firm’s sustainability performance statistically, a new and effective fuzzy logic tool is created. Evolution and assessment are performed by a novel interval type-2 fuzzy logic inference system. The judgment of the inference system is carried out on the basis of type-2 fuzzy logic (T2FL), principal component analysis (PCA), and statistical data analysis. The main input variables include corporate environmental performance (CEP) and corporate financial performance (CFP). The suggested approach can effectively examine a corporation’s sustainable performance, according to experimental findings. A unique approach that makes use of language variables and if-then logic to assist quantitative business sustainability events is the link between CEP and CFP. The recommended test will provide senior administrative leaders with useful information to supervise natural concerns correctly and gauge their commitment to company success.
1. Introduction
The nature of CSP is generally enormously nonlinear, partially inconsistent and multidimensional, vague, and uncertain [1, 2]. A large number of operators, each competent to forecast the extent of business responses to environmental challenges, define the CSP. The main purpose of the CSP is to present an imprecise or uncertain model to evaluate a CEP and CFP. Many studies have recently documented and produced sustainable performance based on two dimensions, which comprise a number of significant affecting factors. There are two dimensions: CEP and CFP. They took some relevant measurements of the financial and environmental markers in order to analyze the two aspects.
The performance assessment has been widely studied. For example in [3], a fuzzy system is suggested to develop the conventional method based on complex proportional assessment approach. A fuzzy model is designed in [4] to construct a model for budget management. In the study of [5], a decision-making system is designed by fuzzy systems, and it is shown that the amusement accuracy of fuzzy approach is better than conventional step-wise weight assessment approach. A fuzzy assessing approach is suggested in [6], for the assessment and selection of fuel vehicles, and it is shown that the fuzzy approach leads to the vehicles with least decreasing carbon emissions. A risk assessment approach is studied in [7], and the risk awareness is classified. In the study of [8], an empirical approach is proposed to analyze the financial scheme of private enterprises in China. The structural equation modeling approach is developed in [9], and the effect if COVID-19 is analyzed.
Creativity is the only way to solve any complex, multilevel problems [10] in every sector or in every domain [11]. Human development is dependent on the psychology of creativity. The employment of creative practices supports the efficacy of a continuous scientific literacy to assist society, and creative thinking influences the growth of cognitive capabilities [11]. Creativity is not synonymous with innovation since it may express in a variety of ways that are unrelated to commercial activity [12]. Although both are required to solve complicated technical, managerial, or social problems [10]. When innovative ideas are applied successfully, they are regarded as the germ of innovation inside a business [13]. A crucial component of innovation is change, as well as novelty. How do companies maintain a competitive edge in environments and markets that change quickly while still promoting long-term growth? One increasingly important way for organizations to achieve this is through sustainability-driven innovation strategies [14, 15].
The literature on the connection between creativity and sustainability has a knowledge gap. This limited understanding of the relationship is caused by the ideas’ ambiguous perception [16]. Moldavska et al. [17] have introduced a special CSA technique for manufacturing industries where corporate sustainability is viewed as a situation and is connected to development goals, modeling the industrial firm utilizing systems representation estimates the evaluation tool, and corporate sustainability is seen as a procedure of constant innovation. Besne et al. [18] Utilize fuzzy logic with LCA to provide effective analytical methodologies. His work includes doing LCAs and cost analyses, normalizing the results of various financial and environmental influencing factors, implementing fuzzy logic by integrating the results, and then computing the fuzzy eco-efficiency index.
The fuzzy systems are extensively employed in different problems such as forecasting [19], controller designing [20], assessing systems [21], sentence representation technique [22], modeling problems [23], mapping problems [24], synchronization systems [25], and many others. However, it is rarely used for sustainability. In this study, a T2FS is developed for attaining corporate sustainability. An IT2FLC is an extensive version of T1-FLS, and it is more accurate and sensitive as reported in literature [26, 27]. Comparative reviews are tabulated in Table 1.
2. Principal Components Analysis
Principle components analysis (PCA) produces an orthogonal transformation from a set of observations of various variables. The order of the principle components is set up such that the first component explains the biggest proportion of variation in data, and the remaining in a decreasing order, explain the remaining proportions of variance. The principle component analysis (PCA) is a statistical method that converts a series of observations of variables that are presumably linked into a set of new values of variables that are linearly uncorrelated, or principal components. The PCA is used to create a low-dimensional representation of a set of data. The most varied linear combinations of variables that are unrelated to one another were presented. In addition to giving extracted parameters for supervised machine learning problems, the PCA may be used to visualize data. The first principal component of a set of features is the normalized linear combination of the features as
It takes the largest variance and it can be calculated by
Let us considered first principal component elements which are the loading of the first principal component and there exists some the principal component loading vectors . The loadings are set up in such a way that the total of their squares equals one. Otherwise, arbitrarily large absolute values for these components may result in an arbitrarily bulky variance.
3. Motivation and Model Formulation
A T2-FLS is proposed to estimated the values of corporate environmental performance (CEP) and CEF. The designed T2-FLS is used to assess the sustainability of corporate. The general scheme is shown Figure 1. The developed model features a multi-variable, nonlinear structure that takes many economic factors into account. CEP and CFP are the two primary inputs of T2-FLS. The T2-FLS rules are optimized in a way that the input variables are used to model the corporate sustainability determinant (CORSUS). A novel method for quantifying CORSUS is produced by the developed rules based on linguistic characteristics. The following is a description of the key variables:(i)WaC: it determines the influence of the company on water resource. Every industry has a limit to use minimum amount of water for product manufacture so that remaining are available for humans and other biotic source.(ii)SS & TM: it measures the pollutant before they are exposed to ecosystem. Lower the discharge higher the sustainable value.(iii)GHG: it comprises of release of GHG like carbon dioxide from various operation. It is essential to minimise the rate of discharge in atmosphere.(iv)VOC: it measures the emission in atmosphere when the body work is painted i.e., body assembly plant.(v)( and ): it measures the emission in atmosphere due to the burning of fossil fuel and from transport.(vi)OIW: it is the solid waste obtain from industry which is landfill. It is essential for a company/industry to minimise the amount of waste landfill rather to reuse or recycle.(vii)HIW: it includes all the waste which is harmful for biotic and abiotic component. It will be more sustainable for company if the amount of production of waste is less. In case of CFP consist of eight parameters which includes:
PCA used the beginning factors to get the projection scores for the CEP and CFP magnitudes. CEP and CFP were valued for the two dimensions using the statistical method PCA by taking into account their initial points in the following equations. CEP is modeled as
To evaluate the number of T2FS must be computing for determining the final indicator in corporate sustainable performance (CORSUS), positive values for and must be allocated which indicating the importance of the two input parameter in the development of the output parameter, CORSUS. The linguistic values are represented as follows. The value 0 is used for very bad, 1 is used for bad, 2 is used for medium, 3 is used for good, and 4 is for very good. For the weight of CEP and CFP, an index have allocated for corporate sustainable performance (CORSUS) of the overall sustainability. This index for CORSUS can be evaluated with the help of the following equation:
When , the minimum value of CORSUS is zero and the maximum value is . To strike a balance between the two variables of sustainability’s environmental and economical components, we have considered the hypothesis. During this spontaneous process in a corporate sustainability, we have computed a new function pattern for the proposed model which is depicted in Figure 2.
4. Solution Procedure
In this section, the fuzzy approach is illustrated. The input variables are CEP and CFP, and the output is CORSUS. The structure is shown in Figure 3. As it is seen for inputs, we have 4 membership functions MFs. The MFs for CEP are denoted by , and for CFP are denoted by . The MFs represent the value of CEP and CFP in various range such as: very-bad (, ), bad (, ), good (, ) and very-good (, ) see Figures 4 and 5. Also, for CORSUS, 6 MFs are considered. The MFs of output are denoted by (see Figure 6). The MFs of output represent the value of CORSUS in various range such as: very-low, low , medium-low, medium , medium-high, and high The details of computations are given as follows. Step 1: get the inputs and , that represent CEP and CFP, respectively. Step 2: compute the upper/lower membership of and as where , , and denote the center of , upper and lower width of , respectively. Similarly, , , and denote the center of , upper and lower width of , respectively. Step 3: the rules are given as follows:1If is and is , Then is 2If is and is , Then is 3If is and is , Then is 4If is and is , Then is 5If is and is , Then is 6If is and is , Then is 7If is and is , Then is 8If is and is , Then is 9If is and is , Then is 10If is and is , Then is 11If is and is , Then is 12If is and is , Then is 13If is and is , Then is 14If is and is , Then is 15If is and is , Then is 16If is and is , Then is 17If is and is , Then is 18If is and is , Then is 19If is and is , Then is 20If is and is , Then is 21If is and is , Then is 22If is and is , Then is 23If is and is , Then is 24If is and is , Then is 25If is and is , Then is If -th MF is fired for CEP, and -th MF is fired for CFP, then the upper/lower rule firing are obtained as Step 4: the CORSUS is computed as where and represent the mean of -th fired MF in output.
5. Results Analysis
We constructed a framework for optimizing CORSUS assessment using an imprecise data set from the field and two input parameters and one output parameter. To make proper decisions, the IT2-FLC was used. Some statistical data analysis was successfully computed to monitor and access the sensitivity and accuracy of the suggested model on. The statistical techniques such as (a) and (b) RMSE have been calculated and compared with the predicted and measured values of flexibility of the proposed model. The value of RMSE is defined by the following mathematical equation:where denotes the actual values of the output, denotes the predicted values of the output, and is the number of the points. In the following (9), we have calculated the determination coefficient as
The average of the squares of the errors can be measured by the value of MAPE. The very smaller values of MAPE make sure better performance. The MAPE has been calculated by using the (10) as
The performance of the proposed models’ efficiency has been assessed using mean absolute error as a benchmark (MAE). The MAE can be determined using the following formula:
Here, we have considered number of data patterns from the industry data set, denotes the predicted value of one data point and denotes the observed value of one data point . The summary of statistical data analysis of output has been given in the Table 2.
6. Simulation Results and Discussion
There are two types of data collected from an industry (CEP and CFP) and we have used the IT2FIS. The IT2FS retrieves these data and, as an output, develops a sustainability performance model based on the data characteristics shown in Figure 7.
6.1. Managerial Inferences
The 2008 financial and economic crisis has led to worries about the impact of CEP and CFP. Banks’ CSR policies have been questioned as a real strategic commitment to the major stakeholders in light of the controversies and scandals surrounding their participation in the crisis. These scandals and controversies have also brought to light shortcomings in several CSP areas. This study demonstrates that merely executing socially and environmentally conscious actions is inadequate to boost financial success; rather, management’s disregard for such practices hurts financial performance. Instead, taking advantage of corporate sustainability performance benefits depends on the industry’s ability to profit from such efforts, since properly managing CORSUS fosters stakeholder linkages that boost performance. For this reason, it is crucial for any industry to grasp that “being excellent” is not enough and that “how they need to be good” by striking the right balance between CEP and CFP is crucial. It is now necessary to have this degree of industrial mindfulness and awareness.
Remark 1. In this study, a type-2 FLS is introduced for attaining corporate sustainability. More recently the new type-3 FLSs have been presented with better ability in practical problems. For the future studies, the suggested approach can be developed by the use of type-3 FLSs, new risk analysis techniques, and new optimization methods [11, 30–37].
7. Conclusions
For the purpose of evaluating corporate sustainable performance (CSP) assessment, we have created and formulated an interval type-2 fuzzy logic control technique. The input parameters (CEP and CFP) and an output parameter were helped to track by the IT2FLC (CORSUS). In order to foresee a range of corporate environmental and financial performance assessment circumstances for an industry, this study contains a safe and reliable approach for drawing conclusions. Because it is more difficult and complex to make judgments from the perspective of the industry, including significant uncertainty, we have taken for the implementation of CSP with the aid of this specific IT2FLC by employing Principal components analysis (PCA). The prediction of different economic, social, and environmental factors aids in CORSUS optimization during an industry’s CSP monitoring system. The authors claim that the recommended IT2FLC strategy is a brand-new and distinctive way to build and regulate the CSP that must be met in order to provide a successful workplace output.
Nomenclature
| WaC: | Water consumption |
| SS: | Suspended solids |
| OIW: | Ordinary industrial waste |
| TM: | Toxic metals |
| GHG: | Greenhouse gas |
| VOC: | Volatile organic compound |
| : | Sulphur dioxide |
| : | Nitrogen oxides |
| ROA: | Return on assets |
| HIW: | Hazardous industrial waste |
| CFP: | Corporate financial performance |
| PE: | Price-earnings |
| ROI: | Return on investment |
| ROE: | Return on equity |
| OM: | operating margin |
| GS: | Growth in sales |
| ROS: | Return on sales |
| DP: | Days payable |
| CEP: | Corporate environmental performance |
| CORSUS: | Corporate sustainable performance |
| CSP: | Corporate sustainability performance |
| EI: | Eco-innovation |
| LCA: | Life Cycle Assessment. |
Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This study was financially supported by the key project of the National Social Science Foundation (Grant no. 18AJY013); the Social Science Program of Hainan Province (Grant no. HNSK(YB)19-10); and the Program of the Ministry of Culture and Tourism (Grant no. TYETP201552).