Assessing Voluntary Guardianship and Personal Autonomy Using a Circular q-Rung Orthopair Fuzzy CoCoFISo Decision Framework

Abstract

A balance between support and independence in guardianship systems is of high concern, especially with those who need help in making decisions. The research presents a novel approach to evaluating voluntary models of guardianship, focusing on the preservation of individual autonomy and examining the underlying decision symmetry in assessing diverse guardianship options. The ultimate solution to the inherent uncertainty and lack of objectivity in expert evaluations is to apply the circular q-rung orthopair fuzzy (Cq-ROF) combined compromise for ideal solution (CoCoFISo) approach, an effective multi-criteria decision-making (MCDM) model that integrates ranking and sorting views using a Cq-ROF framework within a symmetry-oriented analytical perspective. These are five major assessment factors: how well autonomy is preserved, legal and ethical adherence, psychological health, social integration aid, and risk prevention. It explores ten alternative approaches to guardianship, ranging from complete legal guardianship to community-based self-management solutions, and the use of technology as an element of support. The suggested approach can facilitate more sophisticated modelling of expert opinions, rather than relying on simplistic and straightforward distinctions and diverse evaluations. The case study results indicate that the hybrid and supported forms of decision-making could offer opportunities to preserve a high degree of personal autonomy while ensuring safety and compliance. The research gives a coherent, adaptable, and explainable approach to managing ethical and policy-level judgment concerning voluntary guardianship systems.

1. Introduction

In contemporary cultures, the dynamics of human rights, dignity, and independence have evolved, highlighting the importance of reconsidering traditional guardianship models. Voluntary guardianship involves structures where individuals have the right to decide or consent to the level and type of assistance they need in personal, financial, or medical decisions. Personal autonomy, in turn, denotes the freedom to self-govern without external interference. The central research problem lies in achieving a balance between assistance and autonomy for vulnerable groups, such as the elderly, disabled, or those experiencing temporary cognitive impairments. This balance is ethically sensitive and practically challenging, thereby motivating the present study.

The analysis of guardianship models encompasses various aspects, including legal protection, psychological well-being, the level of autonomy to be maintained, social integration, and risk prevention. The qualitative and often ambiguous nature of such criteria is not well captured by traditional evaluation methods. Guardianship, in fact, exists on a continuum; it is not usually a clear-cut black-or-white issue, but may range from complete legal intervention over a person to less modularized approaches such as shared decision-making or technology-assisted independence. Conventional evaluation methods cannot adequately handle the complexity, ambiguity, and subjectivity of guardianship decisions, establishing a clear research gap. Some form of decision-making process is necessary, and this approach should not only evaluate the various models available but also incorporate the symmetry in subjective judgments of stakeholders such as legal professionals, psychologists, relatives, and the individuals themselves. This need underpins the use of the MCDM approach in assessing voluntary guardianship and personal autonomy.

The MCDM approaches offer an analytical framework that presents a more effective way of thinking and assessing various complex alternatives, which are typically confronted by several often-incongruent criteria. Within the framework of the voluntary guardianship, MCDM enables the parallel comparison of the various guardianship models (based on both quantitative and qualitative indicators). The methods are especially applicable when the judgment of experts is conflicted or in cases where a single best solution does not exist and a trade-off between conflicting values must be made (e.g., autonomy versus safety). Nevertheless, human evaluations in these fields are never entirely sure, precise, or definitive. This highlights the need for uncertainty modeling, which motivates the integration of fuzzy extensions.

To overcome this uncertainty, Zadeh [1] developed the fuzzy set theory, which allows for partial membership degree (MD). This concept is expressed in the unit interval of $0$ to $1$ and is used to model vague information. Despite its innovation, fuzzy sets fail in that they are restricted in representing the extent of non-membership fully. Away to overcome this, Atanassov [2] gave an expansion of the concept to intuitionistic fuzzy sets (IFSs) where, in addition to the MD, the non-membership degree (NMD) is used, being able to have a more exact representation of what is uncertain. Extending this further, Yager [3] introduced Pythagorean fuzzy sets (PyFSs), which further relax the requirement that the sum of the squares of the MD and NMD should be one. Yager [4] later on came up with q-rung orthopair fuzzy sets (q-ROFSs), which are even more lenient, because the $q^{th}$ power of MD and NMD must not be greater than $1$. Such progressive extensions of fuzzy sets have turned out to be very effective in dealing with complex decision-making situations characterized by deep uncertainty, as one would find it in the case of guardianship frameworks.

With the requirement to have a superior decision-support model that concomitantly manages multiple criteria, uncertainty, as well as the necessity of sorting and ordering of alternatives, the current research utilizes the CoCoFISo solution in the circular q-rung orthopair fuzzy set (Cq-ROFS) [5] framework. The CoCoFISo, an emerging technique in the field of MCDM, aims to combine the strengths of compromise methods with sorting methods to generate more trustworthy and explainable decisions. It includes ranking consistency, assigning categories, and compromise-based decision logic. The incorporation of this model into the Cq-ROF structure makes the framework able to take into consideration the non-trivial behavior of judgment through hesitation, contradiction, and circular preference dynamics, thus being significantly versatile in its application to the domain of the subtle activity of judging the voluntary guardianship of a person and their autonomy.

To test the proposed model, this study implements it in a real-world setting by considering a scenario where ten voluntary guardianship models are evaluated against five key criteria. These include full legal guardianship models, supported decision-making models, hybrid models, and community-based support systems. The criteria, such as autonomy preservation, legal compliance, ethical compliance, psychological well-being, social integration, and safety assurance, are assessed based on expert opinions. Comparing this method with others in MCDM and conducting a sensitivity analysis demonstrates its robustness, consistency, and reliability. This comprehensive analysis provides a clearer understanding of which guardianship types align most closely with ethical and person-centered decision-making paradigms.

In brief, this research provides an organized decision-support framework, employing the Cq-ROF-CoCoFISo strategy to evaluate voluntary guardianship and personal autonomy. This study is practical both methodologically and practically, as it considers the shortcomings of established approaches and adopts the latest advances in fuzzy set theory and MCDM. Additionally, the model not only facilitates the comparison of complex guardianship alternatives but also enables ethically acceptable decisions based on the dignity of the individual.

1.1. Assessing Voluntary Guardianship and Personal Autonomy

The main tenet of contemporary guardianship policy is the concept of voluntary guardianship and personal autonomy, which strikes a delicate balance between the right to protection and the necessity of self-determination. This assessment of these frameworks requires a person-centered approach that is both careful and respectful of dignity, yet also acknowledges vulnerability. It is stated that a theory-informed guardianship assessment framework should ensure safeguarding autonomy in older adults by also considering the risks of potential abuse and enhancing dignity through the optimal structure of the law [6]. This kind of support for the maximization of voluntary guardianship as part of civil law in a jurisdiction like China demonstrates the advantages of such a model and the importance of respecting the person’s autonomy to make their own decisions, as long as systems are reinforced by clear regulations and institutionalization [7]. Another critical area where guardianship and autonomy intersect concerns the field of medical decision-making. It has been claimed that the adult guardianship system must incorporate the element of medical self-determination through the use of professional proxies and standard forms of appraisal, ensuring that the autonomy of the individual, especially in the health context, is not jeopardized or usurped by outside parties [8]. Simultaneously, psychological and behavioral factors like problem-solving ability, self-motivation, and group decision-making attendance are major predictors of individual autonomy [9], pointing out that autonomy is not only a legal term but also a reflection of cognition and emotion.

Additionally, individual freedom has broader social benefits. According to [10], autonomy is beneficial to social trust, and the socio-economic factors such as homeownership and income affect this deprivation. This suggests that decisions made during guardianship should not be specific to the individual context, but rather address broader societal issues and relationships. Nevertheless, there are still difficulties, and they are especially pronounced when mental health legislation interacts with guardianship laws. Contradiction matters, like that of involuntary mental health guardianship, seem to be more reason to reform an autonomy-giving process with the required protections [11]. Given these multidimensional issues, a powerful, methodological design is suggested in this study that reflects voluntary guardianship and personal autonomy through a Cq-ROF-CoCoFISo decision model. In contrast to conventional evaluation methods, our model is flexible and inclusive, encompassing diverse cognitive, legal, social, and health-related aspects, integrated into an articulated MCDM system that represents a multidimensional, agile assessment process. That way, the choice will be made with a healthy understanding of the independence of the given person, while simultaneously preventing the risk of gaps in the literature and addressing the demands for deeper, more flexible, ethical, and mathematically sound guardianship assessments.

1.2. CoCoFISo MCDM Method

The CoCoFISo is one such strong and dynamic MCDM approach that aims to enhance the quality of decision-making in complex and uncertain environments. It combines the advantages of compromise-based approaches and the classical CoCoSo approach, offering highly versatile and dynamic evaluation in a vast number of applications [12]. Initially formulated to perfect decision algorithms with the aid of ideal solution proximity and compromise programming, CoCoFISo has demonstrated strong results in multi-dimensional analyses, e.g., socio-economic region evaluation [13] and sustainability-driven logistics decision-making based on the q-ROFS [14]. The technique is highly flexible, as it is combined with fuzzy logic systems. It incorporates both ranking and sorting mechanisms, handles diverse weight distributions among decision-makers, and adapts to circular preferences and hesitation in judgments, making it exceptionally robust in uncertain environments. As an illustration, its fuzzy extension picture fuzzy (PF-CoCoFISo) was found to be more optimized when it comes to making decisions in situations that demand sensitive, uncertainty-based judgments [15]. Furthermore, CoCoFISo can accommodate group decision-making, scale effectively to large datasets, and integrate aggregation operators to reflect complex interdependencies among criteria, increasing its practical applicability in real-world scenarios. The additional streamlining of data-focused apps, including the identification of powerful nodes in networks, demonstrates that CoCoFISo can provide support for complementary decision-making characteristics [16]. Consequently, these strengths predispose CoCoFISo to specific areas of subjectivity and ethical consideration, such as voluntary guardianship and a person’s autonomy, where several qualitative and non-determinant variables must be weighed against each other. The versatility of this approach is particularly applicable to our study, where the Cq-ROF-CoCoFISo structure leverages the advantage, providing a formal and practical yet human-tolerant laboratory that facilitates evaluating the options for guardianship based on uncertainty.

1.3. Research Gap and Motivations

The design of guardianship assessment often relies on a legal framework that is too rigid to fully consider the complexity of individual autonomy, especially in voluntary guardianship scenarios. Most methods overlook multi-dimensional criteria and tend to be inflexible when dealing with uncertainty in expert judgments. The reluctance and imprecise principles, along with ethical concerns inherent in guardianship models, are rarely compatible with standard MCDM techniques used in related fields. Additionally, the frameworks that have been agreed upon, such as classical fuzzy sets and IFS, can only represent a limited degree of uncertainty. The Cq-ROFS theory can be applied in this case, as it allows for greater freedom when assigning MD and NMD, and thus, it captures the complexities of hesitation and circular dependencies more effectively than classical fuzzy sets, IFS, and even q-ROFS. This added flexibility lends itself especially well to modelling subjective and ambiguous expert judgments in guardianship assessment.

CoCoFISo is selected due to its exceptional combination of ranking and sorting properties, its incorporation of consensus-based compromise alternatives, and its ability to be robust against a diverse range of weight distributions of involved decision-makers, which other methods, such as TOPSIS, VIKOR, and EDAS, are not able to offer concomitantly. Cq-ROFS and CoCoFISo are a theoretically valuable and practically versatile framework that is able to complement the ethics, laws, and social aspects of voluntary guardianship decision-making. Despite these advancements, a noticeable shortfall remains in implementing advanced fuzzy decision-making tools for evaluating diverse guardianship models against complex standards. Therefore, the present research aims to introduce a new combination of the Cq-ROF environment and CoCoFISo solution to offer a more detailed, ethically sound, and technically viable decision-making framework for testing voluntary guardianship and personal self-control.

1.4. Objectives and Contributions of the Study

This paper aims to develop an intelligent, ethically focused decision support framework and assess voluntary guardianship and personal autonomy using the Cq-ROF CoCoFISo method. The objective is to address the weaknesses of existing assessment models and incorporate advanced fuzzy logic to manage higher levels of uncertainty, improving interpretation across multiple, often conflicting criteria. The research is valuable because it contributes to applying this robust MCDM technique in a socially sensitive domain. It presents five expert-crafted criteria for reviewing ten variant guardianship models. Furthermore, it offers a systematic tool that practitioners and policymakers can use to balance independence with support needs in guardianship assessments. Comparative and sensitivity analyses are included to demonstrate the reliability and versatility of the proposed approach, making a meaningful contribution to both scholarly and practical discussions on guardianship and autonomy.

1.5. Structure of the Study

The rest of the paper is structured as follows: Section 2 covers the fundamentals of Cq-ROFSs; Section 3 outlines the principles of the proposed Cq-ROF-CoCoFISo method; Section 4 features a real case study and discusses its findings. Section 5 provides a comparative assessment and sensitivity analysis, including managerial implications, while Section 6 describes future research directions.

2. Preliminaries

An extension called Cq-ROFS, based on q-ROFS, was created by Yager [5]. This section provides a detailed overview of basic ideas related to qROFS, Cq-ROFS, and their operational laws and score functions. In this context, the hesitancy degree (HD) quantifies the degree of indeterminacy that remains after assigning MD and NMD. Similarly, the circular radius $\overline{\underset{\dot{}}{\mathrm{r}}}\in [0,1]$ and is used to capture additional uncertainty in the Cq-ROFS enhancing their ability to model vagueness in decision-making environments.

Definition 1 

([4]). A qROFS $\mathcal{I}$ in the universe of discourse ${\rm K}$ is defined in the form:

$$\mathcal{I}=\left\{\left(\mathcal{y},\left(\mathfrak{m}\left(\mathcal{y}\right),\mathfrak{n}\left(\mathcal{y}\right)\right)\right)|\mathcal{y}\in {\rm K}\right\}$$

where $\mathfrak{m}:{\rm K}\to \left[\mathrm{0,1}\right]$ and $\mathfrak{n}:{\rm K}\to \left[\mathrm{0,1}\right]$ denote the MD and NMD, respectively. The pair $\left(\mathfrak{m}\left(\mathcal{y}\right),\mathfrak{n}\left(\mathcal{y}\right)\right)$ satisfying the given condition:

$$0\le {\left(\mathfrak{m}\left(\mathcal{y}\right)\right)}^q+{\left(\mathfrak{n}\left(\mathcal{y}\right)\right)}^q\le 1$$

It is known as a q-rung orthopair fuzzy value (q-ROFV). Further, the HD $h$ of $\mathcal{y}\in {\rm K}$ is given as:

$$h\left(\mathcal{y}\right)=\sqrt[q]{1-{\left(\mathfrak{m}\left(\mathcal{y}\right)\right)}^q-{\left(\mathfrak{n}\left(\mathcal{y}\right)\right)}^q}.$$

Definition 2 

([5]). A Cq-ROFS, denoted as $\mathcal{I}$ on a universal set ${\rm K}$, is defined by $\mathcal{I}=\left\{\left(\mathcal{y},\mathfrak{m}\left(\mathcal{y}\right),\mathfrak{n}\left(\mathcal{y}\right),\overline{\underset{\dot{}}{\mathrm{r}}}(\mathcal{y})\right):\mathcal{y}\in X\right\}$, where $\mathfrak{m}\left(\mathcal{y}\right)$ and $\mathfrak{n}\left(\mathcal{y}\right)$ denotes the MD and NMD of the element $\mathcal{y}\in {\rm K}$ to $\mathcal{I}$, respectively. The condition is satisfied by these functions, which operate within the range [0, 1]:

$$0\le {\left(\mathfrak{m}\left(\mathcal{y}\right)\right)}^{\mathrm{q}}+{\left(\mathfrak{n}\left(\mathcal{y}\right)\right)}^{\mathrm{q}}\le 1$$

Also, the HD is defined as:

$$h\left(\mathcal{y}\right)=\sqrt[q]{1-{\left(\mathfrak{m}\left(\mathcal{y}\right)\right)}^q-{\left(\mathfrak{n}\left(\mathcal{y}\right)\right)}^q}$$

and $\overline{\underset{\dot{}}{\mathrm{r}}}\in [0,1]$ be the radius of the circle, which incorporates an additional layer of flexibility in modeling circular uncertainty.

Definition 3 

([5]). For any two circular q-rung orthopair fuzzy values (Cq-ROFVs), we addressed a variety of algebraic laws, including:

• ${\mathcal{I}}_1{\oplus }_{\mathrm{T}\mathrm{N}}{\mathcal{I}}_2=\left(\begin{array}{c}\sqrt[q]{{\left({\mathfrak{m}}_1\left(\mathcal{y}\right)\right)}^{\mathrm{q}}+{\left({\mathfrak{m}}_2\left(\mathcal{y}\right)\right)}^{\mathrm{q}}-{\left({\mathfrak{m}}_1\left(\mathcal{y}\right)\right)}^{\mathrm{q}}{\left({\mathfrak{m}}_2\left(\mathcal{y}\right)\right)}^{\mathrm{q}}},\\ {\mathfrak{n}}_1\left(\mathcal{y}\right){\mathfrak{n}}_2\left(\mathcal{y}\right),\\ \sqrt[q]{{\left({\overline{\underset{\dot{}}{\mathrm{r}}}}_1\left(\mathcal{y}\right)\right)}^{\mathrm{q}}+{\left({\overline{\underset{\dot{}}{\mathrm{r}}}}_2\left(\mathcal{y}\right)\right)}^{\mathrm{q}}-{\left({\overline{\underset{\dot{}}{\mathrm{r}}}}_1\left(\mathcal{y}\right)\right)}^{\mathrm{q}}{\left({\overline{\underset{\dot{}}{\mathrm{r}}}}_2\left(\mathcal{y}\right)\right)}^{\mathrm{q}}}\end{array}\right)$
• ${\mathcal{I}}_1{\oplus }_{\mathrm{T}\mathrm{C}\mathrm{N}}{\mathcal{I}}_2=\left(\begin{array}{c}\sqrt[q]{{\left({\mathfrak{m}}_1\left(\mathcal{y}\right)\right)}^{\mathrm{q}}+{\left({\mathfrak{m}}_2\left(\mathcal{y}\right)\right)}^{\mathrm{q}}-{\left({\mathfrak{m}}_1\left(\mathcal{y}\right)\right)}^{\mathrm{q}}{\left({\mathfrak{m}}_2\left(\mathcal{y}\right)\right)}^{\mathrm{q}}},\\ {\mathfrak{n}}_1\left(\mathcal{y}\right){\mathfrak{n}}_2\left(\mathcal{y}\right),\\ {\overline{\underset{\dot{}}{\mathrm{r}}}}_1\left(\mathcal{y}\right){\overline{\underset{\dot{}}{\mathrm{r}}}}_2\left(\mathcal{y}\right)\end{array}\right)$
• ${\mathcal{I}}_1{\otimes }_{\mathrm{T}\mathrm{N}}{\mathcal{I}}_2=\left(\begin{array}{c}{\mathfrak{m}}_1\left(\mathcal{y}\right){\mathfrak{m}}_2\left(\mathcal{y}\right),\\ \sqrt[q]{{\left({\mathfrak{n}}_1\left(\mathcal{y}\right)\right)}^{\mathrm{q}}+{\left({\mathfrak{n}}_2\left(\mathcal{y}\right)\right)}^{\mathrm{q}}-{\left({\mathfrak{n}}_1\left(\mathcal{y}\right)\right)}^{\mathrm{q}}{\left({\mathfrak{n}}_2\left(\mathcal{y}\right)\right)}^{\mathrm{q}}},\\ {\overline{\underset{\dot{}}{\mathrm{r}}}}_1\left(\mathcal{y}\right){\overline{\underset{\dot{}}{\mathrm{r}}}}_2\left(\mathcal{y}\right)\end{array}\right)$
• ${\mathcal{I}}_1{\otimes }_{\mathrm{T}\mathrm{C}\mathrm{N}}{\mathcal{I}}_2=\left(\begin{array}{c}{\mathfrak{m}}_1\left(\mathcal{y}\right){\mathfrak{m}}_2\left(\mathcal{y}\right),\\ \sqrt[q]{{\left({\mathfrak{n}}_1\left(\mathcal{y}\right)\right)}^{\mathrm{q}}+{\left({\mathfrak{n}}_2\left(\mathcal{y}\right)\right)}^{\mathrm{q}}-{\left({\mathfrak{n}}_1\left(\mathcal{y}\right)\right)}^{\mathrm{q}}{\left({\mathfrak{n}}_2\left(\mathcal{y}\right)\right)}^{\mathrm{q}}},\\ \sqrt[q]{{\left({\overline{\underset{\dot{}}{\mathrm{r}}}}_1\left(\mathcal{y}\right)\right)}^{\mathrm{q}}+{\left({\overline{\underset{\dot{}}{\mathrm{r}}}}_2\left(\mathcal{y}\right)\right)}^{\mathrm{q}}-{\left({\overline{\underset{\dot{}}{\mathrm{r}}}}_1\left(\mathcal{y}\right)\right)}^{\mathrm{q}}{\left({\overline{\underset{\dot{}}{\mathrm{r}}}}_2\left(\mathcal{y}\right)\right)}^{\mathrm{q}}}\end{array}\right)$
• $\mathsf{\eta }{\mathcal{I}}_{\mathrm{T}\mathrm{N}}=\left(\sqrt[q]{1-{\left(1-{\left(\mathfrak{m}\left(\mathcal{y}\right)\right)}^{\mathrm{q}}\right)}^{\mathsf{\eta }}},{\left(\mathfrak{n}\left(\mathcal{y}\right)\right)}^{\mathsf{\eta }},\sqrt[q]{1-{\left(1-{\left(\overline{\underset{\dot{}}{\mathrm{r}}}\left(\mathcal{y}\right)\right)}^{\mathrm{q}}\right)}^{\mathsf{\eta }}}\right),\eta >0$ be any scalar number.
• $\mathsf{\eta }{\mathcal{I}}_{\mathrm{T}\mathrm{C}\mathrm{N}}=\left(\sqrt[q]{1-{\left(1-{\left(\mathfrak{m}\left(\mathcal{y}\right)\right)}^{\mathrm{q}}\right)}^{\mathsf{\eta }}},{\left(\mathfrak{n}\left(\mathcal{y}\right)\right)}^{\mathsf{\eta }},{\left(\overline{\underset{\dot{}}{\mathrm{r}}}\left(\mathcal{y}\right)\right)}^{\mathsf{\eta }}\right)$
• ${\mathcal{I}}_{\mathrm{T}\mathrm{N}}^{\mathsf{\eta }}=\left({\left(\mathfrak{m}\left(\mathcal{y}\right)\right)}^{\mathsf{\eta }},\sqrt[q]{1-{\left(1-{\left(\mathfrak{n}\left(\mathcal{y}\right)\right)}^{\mathrm{q}}\right)}^{\mathsf{\eta }}},{\left(\overline{\underset{\dot{}}{\mathrm{r}}}\left(\mathcal{y}\right)\right)}^{\mathsf{\eta }}\right)$
• ${\mathcal{I}}_{\mathrm{T}\mathrm{C}\mathrm{N}}^{\mathsf{\eta }}=\left({\left(\mathfrak{m}\left(\mathcal{y}\right)\right)}^{\mathsf{\eta }},\sqrt[q]{1-{\left(1-{\left(\mathfrak{n}\left(\mathcal{y}\right)\right)}^{\mathrm{q}}\right)}^{\mathsf{\eta }}},\sqrt[q]{1-{\left(1-{\left(\overline{\underset{\dot{}}{\mathrm{r}}}\left(\mathcal{y}\right)\right)}^{\mathrm{q}}\right)}^{\mathsf{\eta }}}\right)$

Definition 4 

([5]). For any Cq-ROFV denoted by $\mathcal{I}=\left({\mathfrak{m}}_{\mathcal{I}}\left(\mathcal{y}\right),{\mathfrak{n}}_{\mathcal{I}}\left(\mathcal{y}\right);\overline{\underset{\dot{}}{\mathrm{r}}}\left(\mathcal{y}\right)\right)$, then score function is defined as follows:

$$S_{\mathcal{I}}\left(\mathcal{y}\right)=\left({\mathfrak{m}}_{\mathcal{I}}^q\left(\mathcal{y}\right)-{\mathfrak{n}}_{\mathcal{I}}^q\left(\mathcal{y}\right)\right)\ast \left(\mathrm{ṝ}\left(\mathcal{y}\right)\right).$$

(1)

where $S_{\mathcal{I}}\left(\mathcal{y}\right)\in [-1,1]$.

The above definitions, operational rules, and notations provide the necessary foundation for applying the Cq-ROF-CoCoFISo framework. By establishing how circular q-rung orthopair fuzzy values are represented, aggregated, and scored, the preliminaries ensure that the subsequent algorithmic steps are clearly grounded in a mathematically rigorous structure. In the next section, these elements are employed in a systematic decision-making procedure, where the Cq-ROF decision matrix is constructed, alternatives are evaluated against multiple criteria, and the CoCoFISo method is applied to obtain a robust and explainable ranking of guardianship models.

3. Circular q-Rung Orthopair Fuzzy CoCoFISo MCDM Methodology

This section outlines the proposed methodology of Cq-ROF-CoCoFISo, which combines the strengths of Cq-ROFSs with the resource-efficient ranking strategy of the CoCoFISo approach. The Cq-ROFS model can expand its capacity to describe complex uncertainty by incorporating a circular radius term, such as MD, NMD, to guide more intelligent impulses based on expert opinion. In this approach, an advanced fuzzy logic will be integrated into the CoCoFISo framework, enhancing flexibility, discrimination ability, and decision reliability in MCDM settings. This methodology is particularly useful in scenarios such as voluntary guardianship and personal autonomy, facilitating in-depth evaluation, group deliberation, and feasibility assessment, while addressing real-world decision-making problems involving ambiguous and non-contradictory information. The algorithm corresponding to this approach is presented below, and the schematic representation of the methodology is shown in Figure 1.

Step 1: Using the experts’ opinions, the decision matrix defines each parameter’s collection of alternatives, parameters, and weight values.

$${\left(\widehat{{\mathcal{I}}_{\mathcal{i}\mathcal{j}}}\right)}_{\mathcal{m}\times \mathcal{n}}=\begin{array}{c}{A}_1\\ A_2\\ ⋮\\ A_m\end{array}\left[\begin{array}{cccc}{C}_1& C_2& \dots & C_n\\ \left({\mathfrak{m}}_{11},{\mathfrak{n}}_{11},{\overline{\underset{\dot{}}{\mathrm{r}}}}_{11}\right)& \left({\mathfrak{m}}_{12},{\mathfrak{n}}_{12},{\overline{\underset{\dot{}}{\mathrm{r}}}}_{12}\right)& \dots & \left({\mathfrak{m}}_{1\mathcal{n}},{\mathfrak{n}}_{1\mathcal{n}},{\overline{\underset{\dot{}}{\mathrm{r}}}}_{1\mathcal{n}}\right)\\ \left({\mathfrak{m}}_{21},{\mathfrak{n}}_{21},{\overline{\underset{\dot{}}{\mathrm{r}}}}_{21}\right)& \left({\mathfrak{m}}_{22},{\mathfrak{n}}_{22},{\overline{\underset{\dot{}}{\mathrm{r}}}}_{22}\right)& \dots & \left({\mathfrak{m}}_{2\mathcal{n}},{\mathfrak{n}}_{2\mathcal{n}},{\overline{\underset{\dot{}}{\mathrm{r}}}}_{2\mathcal{n}}\right)\\ ⋮& ⋮& \ddots & ⋮\\ \left({\mathfrak{m}}_{\mathcal{m}1},{\mathfrak{n}}_{\mathcal{m}1},{\overline{\underset{\dot{}}{\mathrm{r}}}}_{\mathcal{m}1}\right)& \left({\mathfrak{m}}_{\mathcal{m}2},{\mathfrak{n}}_{\mathcal{m}2},{\overline{\underset{\dot{}}{\mathrm{r}}}}_{\mathcal{m}2}\right)& \dots & \left({\mathfrak{m}}_{\mathcal{m}\mathcal{n}},{\mathfrak{n}}_{\mathcal{m}\mathcal{n}},{\overline{\underset{\dot{}}{\mathrm{r}}}}_{\mathcal{m}\mathcal{n}}\right)\end{array}\right]$$

Step 2: To assess the optimal solution, use the following formula to normalize the alternative matrix according to the cost type:

$$N_{ij}=\left({\displaystyle \frac{{\mathfrak{m}}_{ij}}{\sqrt{\sum _{i=1}^{\mathfrak{m}}{\left({\mathfrak{m}}_{ij}\right)}^2}}},{\displaystyle \frac{{\mathfrak{n}}_{ij}}{\sqrt{\sum _{i=1}^{\mathfrak{m}}{\left({\mathfrak{n}}_{ij}\right)}^2}}},{\displaystyle \frac{{\overline{\underset{\dot{}}{\mathrm{r}}}}_{ij}}{\sqrt{\sum _{i=1}^{\mathfrak{m}}{\left({\overline{\underset{\dot{}}{\mathrm{r}}}}_{ij}\right)}^2}}}\right)$$

(2)

Step 3: This method identified two ways to aggregate the parameter’s weight values during the decision-making process: the sum of the normalized matrix’s power weights ${\mathtt{P}}_i$ and the sum of the normalized matrix’s product by weight values ${\mathtt{S}}_i$.

$${\mathtt{S}}_i=\left(\begin{array}{c}\sum _{j=1}^{\mathfrak{n}}{w}_j{\mathfrak{m}}_{ij},\sum _{j=1}^{\mathfrak{n}}{w}_j{\mathfrak{n}}_{ij},\\ \sum _{j=1}^{\mathfrak{n}}{w}_j{\overline{\underset{\dot{}}{\mathrm{r}}}}_{ij}\end{array}\right);{\mathtt{P}}_i=\left(\begin{array}{c}\sum _{j=1}^{\mathfrak{n}}{\left({\mathfrak{m}}_{ij}\right)}^{w_j},\sum _{j=1}^{\mathfrak{n}}{\left({\mathfrak{n}}_{ij}\right)}^{w_j},\\ \sum _{j=1}^{\mathfrak{n}}{\left({\overline{\underset{\dot{}}{\mathrm{r}}}}_{ij}\right)}^{w_j}\end{array}\right)$$

(3)

The gray relational generation strategy is used to get the ${\mathtt{S}}_i$ value, while the weighted aggregated sum product assessment (WASPAS) multiplication approach is used to get the ${\mathtt{P}}_i$ value.

Step 4: Compute the $\mathtt{S}$ and $\mathtt{P}$ values by weighing the relative importance of all alternatives according to the three evaluation score methodologies, which are described below:

$${Ϗ}_{\mathrm{ia}}={\displaystyle \frac{{\mathtt{P}}_i+{\mathtt{S}}_i}{\sum _{i=1}^{\mathfrak{m}}\left({\mathtt{P}}_i+{\mathtt{S}}_i\right)}}$$

(4)

$${Ϗ}_{ib}={\displaystyle \frac{{\mathtt{P}}_i+{\mathtt{S}}_i}{1+\frac{{\mathtt{P}}_i}{{1+\mathtt{P}}_i}+\frac{{\mathtt{S}}_i}{{1+\mathtt{S}}_i}}}$$

(5)

$${Ϗ}_{ic}={\displaystyle \frac{\mathrm{Ɣ}\left({\mathtt{S}}_i\right)+\left(1-\mathrm{Ɣ}\right)\left({\mathtt{P}}_i\right)}{\left(\mathrm{Ɣ}\underset{i}{\mathrm{m}\mathrm{a}\mathrm{x}}{\mathtt{S}}_i+\left(1-\mathrm{Ɣ}\right)\underset{i}{\mathrm{m}\mathrm{a}\mathrm{x}}{\mathtt{P}}_i\right)}};0\le \mathrm{Ɣ}\le 1$$

(6)

Here,

${Ϗ}_{\mathrm{ia}}$: The weighted sum method (WSM) and weighted product method (WPM) score the arithmetic mean.

${Ϗ}_{ib}:$ Sum of the WSM and WPM relative scores

${Ϗ}_{ic}:$ Balanced compromise between the scores of the WSM and WPM models.

Step 5: The score function (Equation (1)) then calculates the values of ${Ϗ}_i$, which are used to rank the alternatives.

$${\mathit{Ϗ}}_{\mathit{i}}=\left({\left({\mathit{Ϗ}}_{\mathit{i}\mathit{a}}\ast {\mathit{Ϗ}}_{\mathit{i}\mathit{b}}\ast {\mathit{Ϗ}}_{\mathit{i}\mathit{c}}\right)}^{{\displaystyle \frac{1}{3}}}+{\displaystyle \frac{1}{3}}\left({\mathit{Ϗ}}_{\mathit{i}\mathit{a}}+{\mathit{Ϗ}}_{\mathit{i}\mathit{b}}+{\mathit{Ϗ}}_{\mathit{i}\mathit{c}}\right)\right)$$

(7)

4. Assessing Voluntary Guardianship and Personal Autonomy

Voluntary guardianship and the ability to make personal choices and decisions are significant concerns in contemporary ethical, legal, and medical decision-making practices. Voluntary guardianship involves a person willingly agreeing to appoint a guardian through mutual consent, allowing them to make certain life decisions on their behalf while still maintaining some level of autonomy and independence. This approach is especially relevant for older individuals, those with mental disabilities, or people who have experienced mental health breakdowns, as finding a careful balance between autonomy and protection is essential and requires careful consideration. To ensure that individuals retain control over their lives as much as possible while being protected from unnecessary risks, a formal, multidimensional assessment procedure is required.

The MCDM has emerged as a tool to address the complexity of making such decisions. It aids in the systematic evaluation and ranking of multiple decision options when faced with several, often conflicting, criteria. In the context of voluntary guardianship, MCDM supports balanced decision-making by considering various factors, including legal rights, psychological well-being, level of independence, and safety. With input from multiple stakeholders—such as healthcare providers, legal advisors, family members, and the participants themselves—MCDM provides a clear and rational method for selecting the appropriate guardianship model.

Although numerous classical and fuzzy-based solving methods have been used to address human decision problems, this study proposes a new experiment using the Cq-ROF-CoCoFISo approach. The approach is particularly suitable for handling uncertain, indeterminate, and ambiguous data, especially in social and ethical judgments. The Cq-ROFS framework offers a more flexible and open representation of expert opinions, allowing the MD and NMD to be based on the unit interval $[0,1]$. CoCoFISo, however, combines similarities in the ranking and sorting rules of compromise, ensuring the final decision aligns closely with the most acceptable compromise.

The situation is such that a government-supported agency is tasked with determining the most suitable form of voluntary guardianship for a group of people aged 65 years and older who have varying levels of cognitive or physical disabilities. The decision contains an assessment of ten potential models of guardianship based on five critical criteria, as determined by the expert opinions of social workers, psychologists, and legal consultants. Specifically, three experts were consulted: one senior social worker with experience in elderly care, one practicing psychologist specializing in cognitive decline, and one legal consultant with expertise in guardianship law. Their evaluations were aggregated to provide a balanced interdisciplinary perspective. It is important to note that, for this study, hypothetical data have been constructed to simulate realistic conditions, ensuring that the proposed methodology can be transparently demonstrated and validated, while actual field applications remain for future work. The determination is founded upon the following five standards, which are vital to the purity and efficiency of voluntary guardianship:

• ${\mathit{C}}_{\mathbf{1}}$: Degree of Autonomy Preservation: An indication of the amount of individual freedom under the guardianship model to which the individual is privy. The higher the scale, the more independent in aspects of making daily and legal decisions.
• ${\mathit{C}}_{\mathbf{2}}$: Legal and Ethical Compliance: Checks the level of conformance of the model with existing rules and moral standards of guardianship, consent, and personal rights.
• ${\mathit{C}}_{\mathbf{3}}$: Psychological Well-being: Evaluates the emotional and psychological well-being of individuals under the proposed guardianship, encompassing factors such as stress reduction, confidence, and dignity.
• ${\mathit{C}}_{\mathbf{4}}$: Social Integration Support: This refers to the model’s ability to maintain the person socially active and integrated in society, without experiencing isolation.
• ${\mathit{C}}_{\mathbf{5}}$: Risk Mitigation and Safety: Indicates the abilities of the model to reduce the chances of people experiencing abuse, neglect, financial exploitation, or being put in jeopardy regarding their health.

Table 1. List of Alternatives.

Alternatives	Guardianship Model Description
${\mathit{A}}_{\mathbf{1}}$	Full Legal Guardianship
${\mathit{A}}_{\mathbf{2}}$	Limited Guardianship with Periodic Review
${\mathit{A}}_{\mathbf{3}}$	Supported Decision-Making (SDM)
${\mathit{A}}_{\mathbf{4}}$	Shared Decision-Making with Family Members
${\mathit{A}}_{\mathbf{5}}$	Professional Guardian with Community Oversight
${\mathit{A}}_{\mathbf{6}}$	Peer-to-Peer Support Model
${\mathit{A}}_{\mathbf{7}}$	Technology-Assisted Self-Management
${\mathit{A}}_{\mathbf{8}}$	Hybrid Model: SDM+Legal Advocate
${\mathit{A}}_{\mathbf{9}}$	No Guardianship-Community Support Only
${\mathit{A}}_{\mathbf{10}}$	Case-by-Case Temporary Guardianship

shows the list of alternatives.

Then, the new methodology, the Cq-ROF-CoCoFISo, is implemented in a methodical process. Obtain a Cq-ROF decision matrix as illustrated by

Table 2. Circular q-Rung Orthopair fuzzy decision matrix.

	${\mathit{C}}_{\mathbf{1}}$			${\mathit{C}}_{\mathbf{2}}$			${\mathit{C}}_{\mathbf{3}}$			${\mathit{C}}_{\mathbf{4}}$			${\mathit{C}}_{\mathbf{5}}$
	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$
${\mathit{A}}_{\mathbf{1}}$	$0.15$	$0.95$	$0.89$	$0.25$	$0.90$	$0.34$	$0.35$	$0.85$	$0.65$	$0.65$	$0.60$	$0.43$	$0.65$	$0.60$	$0.43$
${\mathit{A}}_{\mathbf{2}}$	$0.70$	$0.50$	$0.55$	$0.90$	$0.44$	$0.87$	$0.35$	$0.85$	$0.65$	$0.55$	$0.70$	$0.23$	$0.65$	$0.60$	$0.43$
${\mathit{A}}_{\mathbf{3}}$	$0.70$	$0.50$	$0.55$	$0.85$	$0.40$	$0.63$	$0.35$	$0.85$	$0.65$	$0.40$	$0.85$	$0.11$	$0.65$	$0.60$	$0.43$
${\mathit{A}}_{\mathbf{4}}$	$0.15$	$0.95$	$0.89$	$0.70$	$0.50$	$0.55$	$0.35$	$0.85$	$0.65$	$0.35$	$0.85$	$0.65$	$0.65$	$0.60$	$0.43$
${\mathit{A}}_{\mathbf{5}}$	$0.55$	$0.70$	$0.23$	$0.65$	$0.60$	$0.43$	$0.15$	$0.95$	$0.89$	$0.25$	$0.90$	$0.34$	$0.40$	$0.85$	$0.11$
${\mathit{A}}_{\mathbf{6}}$	$0.70$	$0.50$	$0.55$	$0.55$	$0.70$	$0.23$	$0.15$	$0.95$	$0.89$	$0.15$	$0.95$	$0.89$	$0.40$	$0.85$	$0.11$
${\mathit{A}}_{\mathbf{7}}$	$0.35$	$0.85$	$0.65$	$0.40$	$0.85$	$0.11$	$0.15$	$0.95$	$0.89$	$0.90$	$0.44$	$0.87$	$0.40$	$0.85$	$0.11$
${\mathit{A}}_{\mathbf{8}}$	$0.35$	$0.85$	$0.65$	$0.35$	$0.85$	$0.65$	$0.90$	$0.44$	$0.87$	$0.85$	$0.40$	$0.63$	$0.70$	$0.50$	$0.55$
${\mathit{A}}_{\mathbf{9}}$	$0.35$	$0.85$	$0.65$	$0.25$	$0.90$	$0.34$	$0.70$	$0.50$	$0.55$	$0.70$	$0.50$	$0.55$	$0.70$	$0.50$	$0.55$
${\mathit{A}}_{\mathbf{10}}$	$0.55$	$0.70$	$0.23$	$0.15$	$0.95$	$0.89$	$0.40$	$0.85$	$0.11$	$0.65$	$0.60$	$0.43$	$0.70$	$0.50$	$0.55$

, with $q=4$. This decision matrix is directly constructed using Cq-ROFVs, which were assigned by three domain experts: a senior social worker, a psychologist, and a legal consultant. Each entry in Table 2 represents the evaluation of one guardianship model under a specific criterion, expressed in terms of Cq-ROF MD, NMD, and radius values. In this way, Table 2 explicitly demonstrates how the proposed framework transforms expert judgments into a structured set of Cq-ROFVs that serve as the computational foundation for further analysis.

The Cq-ROF decision matrix is normalized by using Equation (2) as shown in

Table 3. Normalization of the circular q-Rung Orthopair fuzzy decision matrix.

	${\mathit{C}}_{\mathbf{1}}$			${\mathit{C}}_{\mathbf{2}}$			${\mathit{C}}_{\mathbf{3}}$			${\mathit{C}}_{\mathbf{4}}$			${\mathit{C}}_{\mathbf{5}}$
	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$
${\mathit{A}}_{\mathbf{1}}$	$0.10$	$0.40$	$0.45$	$0.14$	$0.39$	$0.19$	$0.25$	$0.33$	$0.29$	$0.35$	$0.27$	$0.24$	$0.34$	$0.29$	$0.33$
${\mathit{A}}_{\mathbf{2}}$	$0.44$	$0.21$	$0.28$	$0.50$	$0.19$	$0.49$	$0.25$	$0.33$	$0.29$	$0.29$	$0.31$	$0.13$	$0.34$	$0.29$	$0.33$
${\mathit{A}}_{\mathbf{3}}$	$0.44$	$0.21$	$0.28$	$0.48$	$0.17$	$0.35$	$0.25$	$0.33$	$0.29$	$0.21$	$0.38$	$0.06$	$0.34$	$0.29$	$0.33$
${\mathit{A}}_{\mathbf{4}}$	$0.10$	$0.40$	$0.45$	$0.39$	$0.21$	$0.31$	$0.25$	$0.33$	$0.29$	$0.19$	$0.38$	$0.36$	$0.34$	$0.29$	$0.33$
${\mathit{A}}_{\mathbf{5}}$	$0.35$	$0.29$	$0.12$	$0.36$	$0.26$	$0.24$	$0.11$	$0.37$	$0.39$	$0.13$	$0.40$	$0.19$	$0.21$	$0.41$	$0.08$
${\mathit{A}}_{\mathbf{6}}$	$0.44$	$0.21$	$0.28$	$0.31$	$0.30$	$0.13$	$0.11$	$0.37$	$0.39$	$0.08$	$0.43$	$0.50$	$0.21$	$0.41$	$0.08$
${\mathit{A}}_{\mathbf{7}}$	$0.22$	$0.36$	$0.33$	$0.22$	$0.37$	$0.06$	$0.11$	$0.37$	$0.39$	$0.48$	$0.20$	$0.48$	$0.21$	$0.41$	$0.08$
${\mathit{A}}_{\mathbf{8}}$	$0.22$	$0.36$	$0.33$	$0.20$	$0.37$	$0.37$	$0.63$	$0.17$	$0.38$	$0.45$	$0.18$	$0.35$	$0.37$	$0.24$	$0.42$
${\mathit{A}}_{\mathbf{9}}$	$0.22$	$0.36$	$0.33$	$0.14$	$0.39$	$0.19$	$0.49$	$0.19$	$0.24$	$0.37$	$0.22$	$0.31$	$0.37$	$0.24$	$0.42$
${\mathit{A}}_{\mathbf{10}}$	$0.35$	$0.29$	$0.12$	$0.08$	$0.41$	$0.50$	$0.28$	$0.33$	$0.05$	$0.35$	$0.27$	$0.24$	$0.37$	$0.24$	$0.42$

.

$$N_{11}={\displaystyle \frac{0.15}{\sqrt{2.49}}},{\displaystyle \frac{0.95}{\sqrt{5.70}}},{\displaystyle \frac{0.89}{\sqrt{3.87}}}=\left(\mathrm{0.10,0.40,0.45}\right)$$

By the assistance of Equation (3), it is possible to know the sum of the weights of the normalized matrix ${\mathtt{P}}_i$ and the product of the values of the weights of the normalized matrix ${\mathtt{S}}_i$, see

Table 4. ${\displaystyle {\mathtt{S}}_i}$ and ${\displaystyle {\mathtt{P}}_i}$ values.

	${\mathtt{S}}_{\mathit{i}}$			${\mathtt{P}}_{\mathit{i}}$
	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$
${\mathit{A}}_{\mathbf{1}}$	0.23	0.34	0.32	3.67	4.01	3.94
${\mathit{A}}_{\mathbf{2}}$	0.36	0.27	0.30	4.06	3.83	3.89
${\mathit{A}}_{\mathbf{3}}$	0.34	0.27	0.27	4.01	3.84	3.77
${\mathit{A}}_{\mathbf{4}}$	0.24	0.33	0.35	3.71	3.99	4.05
${\mathit{A}}_{\mathbf{5}}$	0.23	0.35	0.20	3.67	4.04	3.56
${\mathit{A}}_{\mathbf{6}}$	0.24	0.34	0.28	3.64	4.01	3.76
${\mathit{A}}_{\mathbf{7}}$	0.23	0.34	0.28	3.69	4.02	3.71
${\mathit{A}}_{\mathbf{8}}$	0.38	0.26	0.37	4.06	3.80	4.10
${\mathit{A}}_{\mathbf{9}}$	0.33	0.28	0.31	3.94	3.85	3.92
${\mathit{A}}_{\mathbf{10}}$	0.30	0.30	0.24	3.86	3.93	3.62

. In this study, the weights of the five criteria were assigned hypothetically by three domain experts (a social worker, a psychologist, and a legal consultant). The selected weight vector is ${\left(0.25,0.15,0.23,0.16,0.21\right)}^T$, which was normalized so that the sum of the weights equals one. These weights reflect a balanced representation of the experts’ judgment across all criteria.

$${\mathtt{S}}_1=\left(\begin{array}{c}\left(\left(0.25\ast 0.10\right)+\left(0.14\ast 0.10\right)+\left(0.25\ast 0.10\right)+\left(0.35\ast 0.10\right)+\left(0.34\ast 0.10\right)\right),\\ \left(\left(0.25\ast 0.40\right)+\left(0.14\ast 0.39\right)+\left(0.25\ast 0.33\right)+\left(0.35\ast 0.27\right)+\left(0.34\ast 0.29\right)\right),\\ \left(\left(0.25\ast 0.45\right)+\left(0.14\ast 0.19\right)+\left(0.25\ast 0.29\right)+\left(0.35\ast 0.24\right)+\left(0.34\ast 0.33\right)\right)\end{array}\right)=\left(\mathrm{0.23,0.34,0.32}\right)$$

$${\mathtt{P}}_1=\left(\begin{array}{c}\left({\left(0.10\right)}^{0.25}+{\left(0.10\right)}^{0.14}+{\left(0.10\right)}^{0.25}+{\left(0.10\right)}^{0.35}+{\left(0.10\right)}^{0.34}\right),\\ \left({\left(0.40\right)}^{0.25}+{\left(0.39\right)}^{0.14}+{\left(0.33\right)}^{0.25}+{\left(0.27\right)}^{0.35}+{\left(0.29\right)}^{0.34}\right),\\ \left({\left(0.45\right)}^{0.25}+{\left(0.19\right)}^{0.14}+{\left(0.29\right)}^{0.25}+{\left(0.24\right)}^{0.35}+{\left(0.33\right)}^{0.34}\right)\end{array}\right)=\left(\mathrm{3.67,4.01,3.94}\right)$$

The ranking of the alternatives is presented in

Table 5. Results of the Cq-ROF-CoCoFISo method.

	$\mathit{Ϗ}\mathit{i}\mathit{a}$			$\mathit{Ϗ}\mathit{i}\mathit{b}$			$\mathit{Ϗ}\mathit{i}\mathit{c}$			$\mathit{Ϗ}\mathit{i}$			Score Values	Ranking
	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$	$\mathfrak{m}$	$\mathfrak{n}$	$\overline{\underset{\dot{}}{\mathbf{r}}}$
${\mathit{A}}_{\mathbf{1}}$	$0.09$	$0.10$	$0.10$	$1.98$	$2.12$	$2.09$	$0.87$	$0.99$	$0.95$	$1.52$	$1.67$	$1.64$	$-0.77$	$10$
${\mathit{A}}_{\mathbf{2}}$	$0.11$	$0.10$	$0.10$	$2.14$	$2.04$	$2.07$	$0.99$	$0.93$	$0.93$	$1.69$	$1.59$	$1.61$	$0.53$	$2$
${\mathit{A}}_{\mathbf{3}}$	$0.11$	$0.10$	$0.10$	$2.12$	$2.05$	$2.02$	$0.98$	$0.93$	$0.90$	$1.67$	$1.60$	$1.56$	$0.37$	$3$
${\mathit{A}}_{\mathbf{4}}$	$0.10$	$0.10$	$0.11$	$1.99$	$2.11$	$2.14$	$0.88$	$0.98$	$0.98$	$1.54$	$1.66$	$1.68$	$-0.64$	$6$
${\mathit{A}}_{\mathbf{5}}$	$0.09$	$0.10$	$0.09$	$1.98$	$2.13$	$1.93$	$0.87$	$1.00$	$0.83$	$1.52$	$1.68$	$1.48$	$-0.74$	$8$
${\mathit{A}}_{\mathbf{6}}$	$0.09$	$0.10$	$0.10$	$1.96$	$2.12$	$2.01$	$0.86$	$0.99$	$0.90$	$1.51$	$1.67$	$1.56$	$-0.76$	$9$
${\mathit{A}}_{\mathbf{7}}$	$0.10$	$0.10$	$0.10$	$1.99$	$2.12$	$1.99$	$0.87$	$1.00$	$0.89$	$1.53$	$1.68$	$1.54$	$-0.71$	$7$
${\mathit{A}}_{\mathbf{8}}$	$0.11$	$0.10$	$0.11$	$2.14$	$2.03$	$2.16$	$1.00$	$0.92$	$1.00$	$1.69$	$1.58$	$1.70$	$0.65$	$1$
${\mathit{A}}_{\mathbf{9}}$	$0.10$	$0.10$	$0.10$	$2.09$	$2.05$	$2.08$	$0.96$	$0.94$	$0.94$	$1.64$	$1.60$	$1.63$	$0.22$	$4$
${\mathit{A}}_{\mathbf{10}}$	$0.10$	$0.10$	$0.09$	$2.05$	$2.09$	$1.95$	$0.93$	$0.96$	$0.85$	$1.61$	$1.64$	$1.50$	$-0.15$	$5$

, which was calculated using Equations (4)–(7), with $\mathrm{Ɣ}=0.6$, and score values in Equation (1).

$$\mathbf{Ϗ}\mathit{i}\mathit{a}={\displaystyle \frac{0.23+3.76}{41.99}}=0.09$$

$${Ϗ}_{ib}={\displaystyle \frac{3.67+0.23}{1+\frac{3.67}{1+3.67}+\frac{0.23}{1+0.23}}}=1.98$$

$${Ϗ}_{ic}={\displaystyle \frac{0.6\left(0.23\right)+\left(1-0.6\right)\left(3.67\right)}{\left(0.6\ast 0.38+\left(1-0.6\right)\ast 4.06\right)}}=0.87$$

$${Ϗ}_i=\left({\left(0.09\ast 1.98\ast 0.87\right)}^{{\displaystyle \frac{1}{3}}}+{\displaystyle \frac{1}{3}}\left(0.09+1.98+0.87\right)\right)=1.52$$

4.1. Result Discussion

The application of the CqROF-CoCoFISo method enabled the detailed assessment of the proposed accidental guardianship models in terms of various evaluation criteria. Out of all of them, $A_8$ (Hybrid Model: SDM + Legal Advocate) was found to be the most suitable method for voluntary guardianship, as Figure 2 demonstrates. This shows a great inclination to the models that are somewhere between personal autonomy and legal protection, i.e., mix of supported decision-making and formal legal advocacy. The second alternative $A_2$ (Limited Guardianship with Periodic Review) was the second-most popular one, which means that the guardianship provision characterized by the possibility of regular review and individual performance is very popular due to its adjustability and the chances to provide personal support. $A_3$ (Supported Decision-Making—SDM) also scored high, highlighting the importance of allowing people to decide for themselves with adequate support.

Conversely, option $A_1$ (Full Legal Guardianship) was the most disfavored option, which was probably because it was very controlling and had little room for individual agency. $A_5$ (Professional Guardian with Community Oversight), $A_6$ (Peer-to-peer Support Model), and $A_7$ (Technology-Assisted Self-Management) were other low-ranked models, which reflected the issue of over-control, inadequate professional capability, or reliance on faceless systems. The existence of moderate support on $A_9$ (No Guardianship Community Support Only) could provide cautionary optimism towards the autonomy-based models held by the community. $A_{10}$ (Case-by-Case Temporary Guardianship) was met with indifference, with a lack of general agreement regarding the soundness of temporary, circumstance-focused systems. $A_4$ (Shared Decision-Making with Family Members) received a low score as well, with the implication that whereas family is essential, teams do not always need full, drawn-out decision-making capability. In summary, the findings highlight the increased interest in dynamic, supportive, and rights-based types of guardianship, particularly those that combine the notions of autonomy and strict legal support. This confirms the quality of the used MCDM methodology in reflecting the intricate specialist opinion concerning such a controversial social system.

4.2. Theoretical Implications

The Cq-ROF-CoCoFISo model advances a broader theory within the domain of decision science and fuzzy logic. It can represent judgment with greater expressive power by combining the circular representation of membership conditions with q-rung orthopair fuzzy logic, thereby offering a more comprehensive approach to explaining human judgment than traditional models. This model not only limits the flexibility of conventional fuzzy settings but also demonstrates the versatility of CoCoFISo in multi-perspective decision-making scenarios. It enhances the understanding of uncertainty and preferences by showing that information on layered uncertainties and preferences can be aggregated in an organized and human-centric manner. Additionally, this research is the first to fully utilize fuzzy MCDM techniques within the field of social policy, thereby narrowing the gap between computational intelligence and human-based governance models.

5. Sensitivity Analysis

To ensure the resilience of the Cq-ROF-CoCoFISo results, we also changed the compromise-weighting parameter $0.1$ and $0.9$. At the end of the spectrum, the overall ranking pattern was remarkably stable: $A_8$ (Hybrid SDM + Legal Advocate) consistently remained the highest ranked condition, and its score continuously increased with the increased focus on compromise effects. $A_2$ (Limited Guardianship with Periodic Review) and $A_3$ (Supported Decision-Making) did not lose their second or third spots either, with only a slight positive drift, or the preference for flexible, review-focused models persisted even in the context of varying decision attitudes. $A_1$ (Full Legal Guardianship) was particularly unpopular at the other end, and its popularity decreased with an increase in $\mathrm{Ɣ}$, which further demonstrates that stakeholders are more punitive when it comes to restrictive arrangements, especially when the model emphasizes a balance between the criteria. These middling choices $A_9$ (Community Support Only), $A_{10}$ (Case-by-Case Temporary Guardianship) showed low sensitivity and never transitioned between the upper and lower clusters, indicating neutrality under a variable weight. On the whole, the trends in the monotonic score plots provided in Figure 3 serve to reiterate that the proposed framework yields stable, plausible rankings and that the suggested hybrid model of guardianship is not a product of a parameter set combination, but rather a robust solution over a reasonable set of parameters.

5.1. Comparison with Existing MCDM Methods

According to

Table 6. Comparison analysis with existing MCDM methods.

Evaluation Criteria	Cq-ROF–CoCoFISo	WASPAS	TOPSIS	VIKOR	EDAS
Handling of Uncertainty	High	Medium	Low	Medium	Medium
Weight Flexibility	High	Medium	Low	Medium	Medium
Ranking Stability	High	Medium	Medium	Low	Medium
Decision-Maker Preference Modeling	High	Medium	Low	Medium	Low
Aggregation Complexity Handling	High	Low	Low	Medium	Medium
Computational Efficiency	Medium	High	High	Medium	Medium
Adaptability to Group Decision-Making	High	Medium	Low	Medium	Medium
Scalability for Large Datasets	High	Medium	Low	Low	Medium
Support for Fuzzy Information	High	Low	Low	Medium	Medium

, the Cq-ROF-CoCoFISo approach outperforms classical multi-criterion decision-making approaches, such as WASPAS [17], TOPSIS [18], VIKOR [19], and EDAS [20], in terms of a broad spectrum of evaluation parameters. The flexibility it enjoys in modeling the preferences of decision-makers, as well as its superiority in dealing with uncertainty through the use of circular q-rung orthopair fuzzy sets, makes it stand out among the rigid conventional models. Moreover, the provision of an adjustable $\mathrm{Ɣ}$ parameter will allow the decision-makers to change the degree of compromise and make the solutions more robust and rank resistant. Group decision-making processes, scaling to large volumes of data, and complex aggregations of fuzzy logic-based functions are also strengths not found in other methods or are poorly represented in them. It is slightly more computationally expensive. Still, the accuracy and adaptability of the framework, along with the quality of the decisions that will be made, render this cost insignificant. The proposed Cq-ROF-CoCoFISo framework will offer a more effective and modern solution to the problem of MCDM under uncertainty. While this analysis provides a comprehensive qualitative comparison, future work will incorporate detailed numerical experiments for the case study to validate further and strengthen the comparative findings.

5.2. Practical and Managerial Implications

The findings of this study provide valuable insights for both practitioners and policymakers involved in guardianship planning, personal autonomy advocacy, and social service delivery. The proposed Cq-ROF–CoCoFISo framework offers a transparent, adaptable, and ethically sensitive decision-support tool that enables stakeholders to evaluate various guardianship models based on multiple, often conflicting, criteria. It is important to note that the current analysis employs hypothetical data to illustrate the methodology; however, the framework is designed to be directly applicable to real-world datasets, allowing for practical implementation in actual decision-making scenarios. Practically, this model can support social workers, legal advisors, and healthcare providers in making informed recommendations that strike a balance between protection and autonomy. From a managerial perspective, institutions and organizations involved in adult care, mental health, or elder services can implement this framework to standardize their decision-making processes while allowing customization based on individual needs and contexts. By integrating expert judgment with fuzzy-based uncertainty handling, the model reduces bias, enhances accountability, and promotes consistent policy development. Overall, this approach strengthens strategic planning and resource allocation while fostering autonomy-respecting guardianship models that incorporate symmetry in ethical and legal considerations.

5.3. Advantages of the Study

This study offers several unique benefits that will contribute both methodologically and practically to the decision-making literature, particularly in the contentious area of voluntary guardianship and personal autonomy. Firstly, this is achieved using the Cq-ROF environment, which easily captures high levels of uncertainty, reluctance, and imprecision commonly associated with expert judgment on social and ethical issues. Secondly, incorporating the CoCoFISo method will enhance the decision-making process by integrating several compromise-based approaches, resulting in a more robust, consistent, and balanced ranking of guardianship models. Furthermore, the introduction of new applications of fuzzy MCDM methods to the social domain expands the use of this popular model into real-life scenarios, thereby broadening the scope of such models beyond industrial or economic fields. The framework is flexible, skill-based, and capable of handling multidimensional data, making it suitable for use by professionals in the legal, healthcare, and policy sectors. Lastly, the systematic organization of sensitivity and comparative analyses supports the future application of the model and enhances its reliability across various contexts, thereby improving its scholarly value and practical relevance.

5.4. Limitations of the Study

This study has some limitations despite its contributions. Firstly, the evaluation relies on experts’ opinions, which, despite careful selection and confirmation, may introduce subjective bias and affect the consistency of ratings across different cultural or legal contexts. Secondly, the Cq-ROF CoCoFISo approach offers advanced uncertainty modeling capabilities; however, because the number of criteria and alternatives influences computational complexity, it may not be highly scalable for very large decision problems. Additionally, the criteria used in this work are not comprehensive and thus fail to capture subtle aspects relevant to guardianship in specific jurisdictions and amidst a changing legislative landscape. Finally, the work focuses on a hypothetical case scenario; however, the real application of such a case would encounter limitations due to data scarcity, resource constraints, or institutional barriers—details that could not be thoroughly addressed in this study.

6. Conclusions

The current study proposes an efficient decision-making model using the Cq-ROF-CoCoFISo approach to evaluate voluntary guardianship and personal autonomy. The proposed technique successfully reflected uncertainty, reluctance, symmetry, and circular correlation in the preferences made by decision-makers because of integrating the flexibility and expressivity of Cq-ROFS into the CoCoFISo method. The application demonstrates that the model is capable of producing a clear and precise ranking of alternatives, highlighting its methodological reliability. The sensitivity analysis confirmed the model’s stability under variations in preference weights. Overall, the study establishes the feasibility and robustness of the Cq-ROF-CoCoFISo framework in addressing complex decision-making scenarios involving the careful assessment of both autonomy and protection.

Future research can be framed towards the flexibility and adaptability of the Cq-ROF-CoCoFISo tool, converting its utilities into other more advanced tools like the spherical fuzzy sets to comprehend the higher levels of uncertainty [21] and interval-valued T-spherical fuzzy information [22], which have been found helpful in handling uncertainties at other higher levels. It can further help in modeling the complex interdependence of criteria, especially in education, healthcare, and industrial sectors, effectively by the inclusion of aggregation mechanisms such as Frank operators [23] and Dombi-based algorithms [24]. In addition, the further extension of the set of applications of circular-based fuzzy systems to a broader range of areas, including the automotive industry [25], might contribute to the more extensive representation of the context-specific scene of the decision-making process, and its consistency with ambiguities and dynamic states that flow in the real world. Moreover, the present model should be tested in a real-world environment to validate its practical effectiveness and adaptability. Additionally, future work will incorporate detailed numerical analyses for the present case study to quantitatively validate and enhance the comparative findings.