Integration of Artificial Intelligence and ARAS Method for Multi-Criteria Decision-Making in Organizational Performance Assessment

Authors

• Suresh Deepak Gurubasannavar Sr. Director Information Technology,Texas,USA

Abstract

This investigation uses an integrated combination of artificial intelligence (AI) and multi-criteria decision-making (MCDM) approaches, specifically the Aggregate Ratio Assessment (ARAS) method, to develop comprehensive decision support frameworks for assessing organizational performance. The study analyses five critical organizational aspects: financial management, customer relations, operational processes, knowledge development, and organizational capability. By using standardized and weighted matrices, the ARAS technique transforms multi-dimensional criteria into uniform metrics suitable for unbiased evaluation. The findings reveal that financial factors demonstrate the greatest local importance (0.3325) and BNP measurement (0.407), which highlight their key influence on decision processes. Knowledge development, which exhibits significant overall importance (0.2793), emerges as essential for sustained organizational success. The approach generates robust rankings through applied operational calculations (Si) and importance parameters (Ki), with the leading factor recording a Ki value of 0.837833. This integrated AI-MCDM method successfully combines mathematical rigor with intelligent computation, delivering reliable and streamlined decision results. These results provide meaningful guidance for resource allocation, strategy formulation, and operational improvement within organizations. This work advances the growing field of AI-enhanced decision support algorithms, demonstrating how conventional MCDM frameworks can be strengthened with AI integration to address complex business constraints across a variety of industries.

Keywords: Artificial Intelligence , Multi-Criteria Decision Making, ARAS Method, Decision Support Systems, Organizational Performance, Normalized Matrix, Weighted Matrix, Utility Function, Expert Systems, Machine Learning.

Artificial intelligence (AI) has become an essential tool in decision support systems due to its ability to identify patterns in large, complex datasets, solve problems quickly, and provide accurate predictions through machine learning [1]. AI-powered systems can automate increasingly complex tasks, leading to improvements in efficiency, accuracy, and innovation across a variety of industries. As AI continues to advance, it is expected to surpass current capabilities in speed, efficiency, and software development, becoming a critical decision support tool in many industries [2]. Decision making is a rational process in which decision makers (DMs) evaluate alternative ways to achieve specific objectives or satisfy stakeholder needs [3]. Multi-criteria decision making (MCDM) is a structured method that has gained prominence in operations research, AI, and computer science. It can handle both explicit and implicit data, making it useful for solving complex problems such as data transformation, weighting, and calculations. Fuzzy MCDM (FMCDM) has emerged as a solution to decision-making challenges, particularly in control engineering, expert systems, AI, and management science [4].AI and MCDM techniques are often combined to handle uncertainty and improve decision-making. Approaches such as the Analytic Hierarchy Process (AHP), the Analytic Network Process (ANP), and the Data Context Analysis (DEA) combine traditional methods with AI to improve decision-making [6]. These combined techniques have been used in areas such as material selection in design, where AI helps identify optimal solutions based on multiple criteria.[7].In addition, the role of AI in automating routine tasks is transforming industries.

The level of automation depends on the level of human involvement, with greater automation reducing the need for human intervention in repetitive tasks, freeing humans to focus on more complex problem-solving tasks [8].The financial crisis exposed the limitations of traditional assessments such as credit ratings, raising concerns about transparency and conflicts of interest. Similarly, environmental, social, and governance (ESG) ratings, which originated in the 1970s, are distinct from credit ratings but share similarities in assessing risk. Both are becoming increasingly relevant in today’s marketplace as businesses and investors seek comprehensive insights into company performance [11].Fuelled by advances in data storage and analytics, the rise of big data is reshaping industries. Initiatives like the Obama administration’s Big Data Research and Development Initiative aim to unlock the value of data by transforming it into actionable insights [12]. Big data drives economic growth, innovation, and scientific research, while also supporting national goals like disaster response and resource management.[13].Along with other technologies such as machine learning and neural networks, it is revolutionizing industries by combining mathematical theory with practical applications, establishing itself as one of the most impactful technological advances in the sector.[15]

Finance:Within the finance sector, it encompasses the practice of money management, which includes activities such as capital allocation, loan acquisition, loan extension, financial planning, wealth accumulation, and financial planning. The financial structure includes cash flow management, investment portfolio oversight, and credit facility provision.

Customer:A customer is generally someone who buys a product, service, or item. More specifically, customers are individuals or businesses who actively purchase, have previously purchased, or are interested in purchasing a product or service from another person or company.

Internal process:Internal processes, also known as business processes or systems, are the practices and structures that form the foundation of an organization. These systems help businesses run smoothly, ensuring that they can produce their products or services effectively and efficiently.

Learning and growth:Growth refers to physical and biological changes, while development involves changes in function and behavior. Learning is the process of adapting to environmental factors, and maturation describes the progression to an adult-like state in terms of skills or behaviors.

Organization competence:Organizational capabilities refer to a company's ability to deliver value and compete successfully in the marketplace. These capabilities are key to a company's success and form the foundation of its products, services, and overall reputation.

Local weights:Local weights are a concept used in areas such as machine learning, statistics, and network routing to give greater importance or influence to nearby or contextually relevant data points and features. In contrast to global weights, which are applied uniformly across an entire dataset or network, local weights are context-specific and vary depending on the data point or option being analyzed.

Overall weights:A weighted average is a statistical technique used to calculate the average of a set of values, in which the individual group averages are adjusted for their respective sample sizes. This method gives increased importance to groups with larger sample sizes, thereby enabling them to exert a more substantial influence on the overall average.

BNP:BNP stands for B-type natriuretic peptide, a hormone derived from the heart, and the term usually refers to a laboratory assay that measures its concentration. Increased BNP concentrations can indicate heart stress, making this assay instrumental in identifying pathological conditions such as heart failure. It helps medical practitioners assess the degree of heart failure and monitor the effectiveness of treatment.

STD BNP:STD BNP stands for Standardized B-type Natriuretic Peptide, which refers to the measurement of BNP levels using a standardized method. This test is used to assess heart function, with high BNP levels indicating possible heart stress or dysfunction. It helps in the diagnosis, evaluation, and management of heart-related conditions

ARAS Method

Multi-criteria decision-making (MCDM) methods find extensive application in various domains of human endeavour. Within an MCDM framework, each alternative is distinguished by multiple criteria that can be qualitative or quantitative [16]. These criteria often have different units of measurement and require optimization along different paths. Normalization procedures are used to convert the criterion values ​​into comparable metrics. [17].This manuscript introduces a novel method called Admixture Ratio Assessment (ARAS). To validate this approach, a practical case study investigating microclimate conditions in office environments is presented. [18].Sustainable development and environmental integrity can be significantly compromised by catastrophic events. A large number of construction activities are implemented through mechanized systems that function as integrated [19].Technological networks. In process engineering, one of the most important considerations includes performance metrics, which are related to the economic gains and losses generated by system operations [21]. To rank alternatives and identify optimal solutions, the novel ARAS method is implemented. A typical MCDM challenge involves prioritizing a limited set of decision alternatives, each characterized by unique decision criteria that require simultaneous evaluation [22].The ARAS algorithm operates on the premise that complex global phenomena can be understood through straightforward comparative analysis [23]. It theorizes that the degree of optimality of an alternative is established by the proportional relationship between the sum of its normalized and weighted criterion values [27].This ratio indicates the proximity of the alternative to the best solution [29]. According to ARAS principles, the utility function value characterizing the comprehensive comparative performance of a particular alternative - the value of which maintains direct proportionality to the total influence of the criterion values ​​and weights relevant to the project under consideration [30].

Table 1.Artificial intelligence expert system

The table presents the importance of different categories (finance, customer, internal process, learning and development, organizational capability) in an AI expert system. Finance has the highest local weight (0.3325) and a strong BNP value (0.407), indicating its important role. Learning and development shows the highest overall weight (0.2793) with stable BNP values. Customer and internal process have moderate weights, while organizational capability has the lowest BNP values, indicating its low relevance in this context.

FIGURE 1.Artificial Intelligence Expert System

Figure 1 presents data for the artificial intelligence expert system in various categories. Finance has the highest local weight (0.3325) and BNP value (0.407), indicating its strong influence. Learning and Development shows the highest overall weight (0.2793) with stable BNP and STD_BNP values. Customer and InternalProcess have moderate weights, while OrganizationalCapability shows the lowest BNP values, indicating its weak fit in the system. The values ​​highlight the varying importance of each category in the expert system analysis.

Table 2. Calculation of maximum value

Table 2 provides a calculation of the peak values ​​across the various factors. The “local weights” and “overall weights” indicate the relative importance of each criterion, while “BNPa” and “STD_BNPb” indicate the specific metrics or ratings associated with each factor. The aggregate figures illustrate the overall weight for each category, with values ​​for distinct dimensions including finance, customer relations, internal operations, learning and development, and organizational capability fluctuating. These calculations make it easier to assess the effectiveness or importance of various organizational components, where higher weights indicate stronger influence or applicability.

Table 3.Normalized Matrix

Table 3 shows the normalized matrix, where “Local Weights” and “Overall Weights” are adjusted values ​​that reflect the importance of each factor. The columns labeled “BNPa” and “STD_BNPb” represent the specific scores for each category. The normalized values ​​across dimensions such as Financial, Customer, Internal Process, Learning & Development, and Organizational Capability allow for a more balanced comparison of the importance of each factor. These normalized figures highlight the relative contributions of each factor to the overall performance metrics.

FIGURE 2. Normalized matrix

Figure 2 presents a normalized matrix that displays the modified values ​​for various factors in four columns. These include standardized metrics for “Local Weights,” “Overall Weights,” “BNPa,” and “STD_BNPb.” It facilitates comparative analysis of factors including Team, Finance, Customer, Internal Process, Learning and Development, and Organizational Capability based on their standardized scores. The values ​​emphasize the relative importance and effectiveness of each factor within a consistent framework, which helps to clearly understand their contributions to the comprehensive analysis.

Table 4. Weighted Normalized Matrix

Table 4 shows a weighted normalized matrix, in which the standardized values ​​of each factor are adjusted according to their respective weight coefficients. The matrix consists of columns labeled “Local Weights,” “Overall Weights,” “BNPa,” and “STD_BNPb.” It provides a weight-adjusted representation of various categories, including financial, customer, internal process, learning & development, and organizational capability. These values ​​demonstrate the importance of each factor after determining its relative importance, thereby facilitating a more efficient assessment of their contributions to the detailed analysis.

FIGURE 3. Weighted Normalized Matrix

Figure 3, which shows the adjusted values ​​for different factors after applying their respective weights. The matrix contains columns for “Local Weights,” “Overall Weights,” “BNPa,” and “STD_BNPb.” It captures weighted values ​​for categories such as Financial, Customer, Internal Process, Learning & Development, and Organizational Capability. These weighted values ​​provide a more refined perspective on the importance of each factor, emphasizing their relative contributions to overall performance or analysis in a standardized manner.

Table 5. Final Result

Table 5 presents the final results, showing the scores (Si), importance (Ki) and associated rankings for the various factors. The values ​​indicate how each factor performs after the analysis, with “Si” representing the score and “Ki” representing its weight or importance. Ranks are assigned based on an overall rating, with the first ranked factor having the highest combined score and importance. The table highlights the relative position of each factor, showing that the first ranked factor has the most significant impact overall.

FIGURE 4. Si and Ki

Figure 4 illustrates the relationship between the scores (Si) assigned to various factors and their associated importance coefficients (Ki). Each set of linked values ​​represents a performance measure and weight parameter for a particular factor. The Si values ​​indicate the operational performance of the factor, whereas the Ki values ​​reflect its relative importance. This comparative analysis helps not only to assess individual factor performance, but also to determine its importance within the overall assessment, thereby informing decision-making processes based on these quantitative indicators.

FIGURE 5. Shows the Rank

Figure 5 presents a hierarchical order of the factors according to their effectiveness and importance. The ranks are assigned in order: Rank 1 represents the highest position, followed by Rank 2, Rank 3, Rank 4, and Rank 5. This hierarchical arrangement demonstrates the relative importance and effectiveness of each factor within the analysis, with Rank 1 representing the most influential factor. This figure illustrates the order of priority, facilitating the prioritization of factors based on their total rating.

1. INTRODUCTION
2. Materials a nd Method
3. Analysis a nd Discussion
4. Conclusion

This study highlights the successful, specifically the ARAS (Associative Ratio Assessment) approach, to build effective decision support systems. It demonstrates how AI-driven expert systems can assess complex organizational factors such as finance, customer relationships, internal processes, learning and development, and organizational capability. The findings show that finance, with a high local weight (0.3325) and BNP value (0.407), plays a key role in decision-making. Learning and development, with a high overall weight (0.2793), is essential for long-term sustainability. To standardize the comparison of factors, this study uses normalized and weighted matrices, which enable objective prioritization. The ARAS method effectively handles complex decision-making situations by transforming multi-dimensional criteria into comparable metrics. The final ranking based on utility values ​​(Si) and importance coefficients (Ki) provides clear guidance for selecting the optimal alternatives, with the top-ranking factor reaching 0.837833 Ki. This research advances AI-assisted MCDM applications, demonstrating how traditional decision-making models can be enhanced by AI. These results have important implications for organizational management, particularly in resource allocation and strategic planning. As AI advances, its integration with MCDM methods such as ARAS is poised to transform decision-making support systems, providing more accurate, efficient, and reliable outcomes across a variety of industries. Future research should explore its broader applications and the potential of emerging AI technologies to further enhance decision-making processes and organizational competitiveness.

REFERENCES

1. .Eldrandaly, Khalid A. "An Intelligent MCDM Approach for Selecting the Suitable Expert System Building Tool." Int. Arab J. Inf. Technol. 4, no. 4 (2007): 365-371.
2. Eldrandaly, Khalid, Abdel Hadi Ahmed, and Nabil AbdelAziz. "An expert system for choosing the suitable MCDM method for solving a spatial decision problem." In 9th International conference on production engineering, design and control, no. February 2009, p. 12. 2009.
3. Wang, Ke, Ziyi Ying, ShankhaShubhraGoswami, Yongsheng Yin, and Yafei Zhao. "Investigating the role of artificial intelligence technologies in the construction industry using a Delphi-ANP-TOPSIS hybrid MCDM concept under a fuzzy environment." Sustainability 15, no. 15 (2023): 11848.
4. Nguyen, Thi Minh Hang, V. P. Nguyen, and D. T. Nguyen. "A new hybrid Pythagorean fuzzy AHP and COCOSO MCDM based approach by adopting artificial intelligence technologies." Journal of experimental & theoretical artificial intelligence 36, no. 7 (2024): 1279-1305.
5. Divya Soundarapandian (2024). Machine Learning-Driven Reliability Engineering for E-Commerce Sites: A Study on Browsing, Cart, and Checkout Phases. Journal of Business Management and Entrepreneurship, 3(1), 1-9. doi: https://doi.org/10.55124/jbme.v3i1.239
6. . Bączkiewicz, Aleksandra, BartłomiejKizielewicz, AndriiShekhovtsov, JarosławWątróbski, Jakub Więckowski, and WojciechSalabun. "Towards an e-commerce recommendation system based on MCDM methods." In 2021 International Conference on Decision Aid Sciences and Application (DASA), pp. 991-996. IEEE, 2021.
7. Sałabun, Wojciech, and Andrzej Piegat. "Comparative analysis of MCDM methods for the assessment of mortality in patients with acute coronary syndrome." Artificial Intelligence Review 48, no. 4 (2017): 557-571.
8. Baydaş, Mahmut, and NazlıErsoy. "Artificial Intelligence-Assisted Multi-Criteria Decision-Making Methodology: From Research Trends to the Future Roadmap." TürkDoğave Fen Dergisi 14, no. 1 (2025): 180-191.
9. Raghavendra Sunku. (2025). Beyond digitalization strategic automation as a driver of policy administration performance using linear and random forest regression. International Journal of Computer Science and Data Engineering, 2(4), 260. https://doi.org/10.55124/csdb.v2i4.260
10. Aljohani, Abeer. "AI-Driven decision-making for personalized elderly care: a fuzzy MCDM-based framework for enhancing treatment recommendations." BMC Medical Informatics and Decision Making 25, no. 1 (2025): 1-16.
11. Frini, Anissa. "A multicriteria intelligence aid methodology using MCDA, artificial intelligence, and fuzzy sets theory." Mathematical Problems in Engineering 2017, no. 1 (2017): 9281321.
12. Gundala, Tirumala Rao. (2024). Performance Optimization for Micro-Frontend-Based Applications: A Predictive Analysis Using XG Boost Regression. Journal of Business Intelligence and Data Analytics, 2(3), 1-6. https://doi.org/10.55124/jbid.v2i3.256
13. Perikala. K (2025). Architecting MCP-Based Platforms for Enterprise-Scale Agentic Generative AI. Journal of Business Intelligence and Data Analytics, 2(3), 1-8. https://doi.org/10.55124/jbid.v2i3.264
14. Lyu, Hai-Min, Zhen-Yu Yin, Annan Zhou, and Shui-Long Shen. "MCDM-based flood risk assessment of metro systems in smart city development: A review." Environmental Impact Assessment Review 101 (2023): 107154.
15. Bahrammirzaee, Arash, Ali RajabzadehGhatari, Parviz Ahmadi, and KuroshMadani. "Hybrid credit ranking intelligent system using expert system and artificial neural networks." Applied Intelligence 34, no. 1 (2011): 28-46.
16. Samuel, K. K. M. (2025). Block chain and AI Convergence in Financial Technology: A WASPAS-Based Analysis. International Journal of Cloud Computing and Supply Chain Management, 1(3), 1-6. doi: https://doi.org/10.55124/ijccscm.v1i3.247
17. Hsueh, Sung-Lin, Yuan Feng, Yue Sun, RuqiJia, and Min-Ren Yan. "Using AI-MCDM model to boost sustainable energy system development: A case study on solar energy and rainwater collection in guangdong province." Sustainability 13, no. 22 (2021): 12505.
18. Taylan, Osman, Bulent Guloglu, Abdulaziz S. Alkabaa, SalihSarp, Hassan M. Alidrisi, Ahmad H. Milyani, HishamAlidrisi, and Mohammed Balubaid. "AI based fuzzy MCDM models: Comparison and evaluation of dissimilar outcomes, an application to enhance pilot recruitment process." Expert Systems 41, no. 9 (2024): e13590.
19. Kornyshova, Elena, and Camille Salinesi. "MCDM techniques selection approaches: state of the art." In 2007 ieee symposium on computational intelligence in multi-criteria decision-making, pp. 22-29. IEEE, 2007.
20. Ali, Rahman, Anwar Hussain, Shah Nazir, Sulaiman Khan, and Habib Ullah Khan. "Intelligent decision support systems—An analysis of machine learning and multicriteria decision-making methods." Applied Sciences 13, no. 22 (2023): 12426.
21. Liu, Nana, and ZeshuiXu. "An overview of ARAS method: Theory development, application extension, and future challenge." International Journal of Intelligent Systems 36, no. 7 (2021): 3524-3565.
22. Zavadskas, EdmundasKazimieras, and ZenonasTurskis. "A new additive ratio assessment (ARAS) method in multicriteria decision‐making." Technological and economic development of economy 16, no. 2 (2010): 159-172.
23. Raj Mishra, A., G. Sisodia, K. Raj Pardasani, and K. Sharma. "Multi-criteria IT personnel selection on intuitionistic fuzzy information measures and ARAS methodology." Iranian Journal of Fuzzy Systems 17, no. 4 (2020): 55-68.
24. Gundala, Tirumala Rao. (2025). Predictive Analytics for SSO Performance: Improving Authentication Response Times in Oracle Enterprise Environments Using Linear Regression, Random Forest Regression. International Journal of Computer Science and Data Engineering, 2(4), 1–6. doi: https://dx.doi.org/10.55124/csdb.v2i4.265
25. Jovčić, Stefan, Vladimir Simić, PetrPrůša, and MomčiloDobrodolac. "Picture fuzzy ARAS method for freight distribution concept selection." Symmetry 12, no. 7 (2020): 1062.
26. Christensen, Thomas J. "China, the US-Japan alliance, and the security dilemma in East Asia." International Security 23, no. 4 (1999): 49-80.
27. Schwartz, Lisa A., Jacob Kampen, and Banco Mundial. Agricultural extension in east Africa. Vol. 164. Washington, DC: World Bank, 1992.
28. Divya Soundarapandian (2025). Machine Learning Algorithms for Optimizing Search Personalization and Site Reliability in E-Commerce Platforms A Comparative Analysis of Linear Regression, SVR, and AdaBoost. International Journal of Artificial intelligence and Machine Learning, 3(3), 1-7. https://doi.org/10.55124/jaim.v3i3.286
29. Aka, V. P. K. (2024). Strategic Framework for SAP S/4HANA Transformation Planning: Support Vector Regression Analysis of Migration Parameters and Implementation Paths. International Journal of Computer Science and Data Engineering, 1(2), 1–7. doi: https://dx.doi.org/10.55124/csdb.v1i2.262
30. Dorfeshan, Yahya, SeyedMeysamMousavi, EdmundasKazimierasZavadskas, and JurgitaAntucheviciene. "A new enhanced ARAS method for critical path selection of engineering projects with interval type-2 fuzzy sets." International Journal of Information Technology & Decision Making 20, no. 01 (2021): 37-65.
31. Karagöz, Selman, MuhammetDeveci, Vladimir Simic, and Nezir Aydin. "Interval type-2 Fuzzy ARAS method for recycling facility location problems." Applied Soft Computing 102 (2021): 107107.
32. Paul, Debapriyo, P. Agarwal, and Shankar Chakraborty. "Performance appraisal of Indian state police forces using ARAS method." Management Science Letters 6, no. 5 (2016): 361-372.
33. Aka, V. P. K. (2025). Enhancing SAP Full-Cycle Automation and Cost Efficiency with OpenText VIM: A Regression-Based Predictive Study. International Journal of Cloud Computing and Supply Chain Management, 1(3). doi: https://doi.org/10.55124/ijccscm.v1i3.247
34. Liu, Nana, and Zeshui Xu. "An overview of ARAS method: Theory development, application extension, and future challenge." International Journal of Intelligent Systems 36, no. 7 (2021): 3524-3565.
35. Zavadskas, EdmundasKazimieras, and ZenonasTurskis. "A new additive ratio assessment (ARAS) method in multicriteria decision‐making." Technological and economic development of economy 16, no. 2 (2010): 159-172.
36. Zavadskas, EdmundasKazimieras, ZenonasTurskis, and TatjanaVilutiene. "Multiple criteria analysis of foundation instalment alternatives by applying Additive Ratio Assessment (ARAS) method." Archives of civil and mechanical engineering 10, no. 3 (2010): 123-141.
37. Turskis, Zenonas, and EdmundasKazimierasZavadskas. "A novel method for multiple criteria analysis: grey additive ratio assessment (ARAS-G) method." Informatica 21, no. 4 (2010): 597-610.
38. Stanujkic, Dragisa, and RodoljubJovanovic. "Measuring a quality of faculty website using ARAS method." (2012).
39. Raghavendra Sunku. (2024). AI-Powered Forecasting and Insights in Big Data Environments. Journal of Business Intelligence and Data Analytics, 1(2), 254. https://doi.org/10.55124/jbid.v1i2.254
40. Adali, EsraAytac, and AyģegülTuşIşık. "Air conditioner selection problem with COPRAS and ARAS methods." MantasSosyalAraştırmalarDergisi 5, no. 2 (2016): 124-138.
41. Perikala. K. (2025). Cloud-Native NoSQL Foundations for Large-Scale Generative AI Platforms. International Journal of Cloud Computing and Supply Chain Management, 1(3), 1-6. doi: https://doi.org/10.55124/ijccscm.v1i3.248
42. Samuel, K. K. M. (2025). Improving Big Data Intelligence Using Entropy Weighted Method for Cloud-Based AutoML Evaluation. Journal of Artificial Intelligence and Machine Learning, 3(3), 1-6. doi: https://doi.org/10.55124/jaim.v3i3.280