Circular Waste Management Through Sustainability Governance in the Energy Industry

The respondent profile in this study provides valuable insights into the demographic characteristics influencing the utilization of FABA at the Asam-Asam coal-fired power plant. The sampling technique employed was purposive sampling, whereby the respondents were selected based on predetermined criteria. The criteria were as follows: (1) community members residing within the “Ring-1” radius of the Asam-Asam power plant who possess basic knowledge of FABA; (2) local government officials involved in monitoring or managing regional development and environmental impacts; (3) external companies and MSMEs with the potential to use FABA as a raw material for their products. Of the 115 respondents, the majority were male (96.5%), comprising 111 individuals, while only four respondents (3.5%) were female. This imbalance reflects the dominance of male participation in technical and construction-related sectors rather than differences in knowledge, and it is acknowledged as a potential source of bias in interpretation. The age distribution was diverse, with an average age of 30.56 years and a range of 18-54 years. This age diversity indicates that FABA utilization encompasses individuals from various age groups.

In quantitative research, validity and reliability are critical aspects that ensure the quality and dependability of a measurement instrument. Validity refers to the extent to which an instrument accurately measures what it is intended to measure, whereas reliability assesses the consistency of the results when the instrument is applied under similar conditions. Instrument validity was tested using the Pearson Product-Moment Correlation technique, which evaluates the strength of the relationship between each indicator and the total score of its construct. Pilot testing. Prior to the main survey, a pilot test was conducted with 29 respondents to evaluate item clarity, preliminary item-total validity, and initial reliability. This pilot size aligns with scale development recommendations suggesting 10-30 participants as an appropriate range for preliminary studies. The pilot findings were used to refine the wording and filter items before the main data collection. Following instrument refinement, data were collected from 115 respondents and analyzed using EFA. This sample size satisfies the general guidelines (≥ 100-200; N : variable ratio of 3-10) and was deemed adequate based on the high KMO index (≈0.94) and significant Bartlett’s test, both indicating the factorability of the correlation matrix. All final reliability reporting (factor-level alpha) and factor interpretation were based on the main survey (n = 115) rather than pilot data. Based on the calculations, the pilot test involved 29 respondents, yielding degrees of freedom (df) = n - 2 = 29 - 2 = 27 for the correlation test. At a significance level of α = 0.05, the critical r-value used as a reference was 0.3673, which was obtained from the Pearson statistical distribution table. The validity test results demonstrated that all indicators, from I1 to I18, had Pearson Correlation values greater than 0.3673. Furthermore, each indicator had a significance value (Sig. 2-tailed) below 0.05, indicating that the correlations between each indicator and the total score were statistically significant (Table 3). Thus, it can be concluded that all indicators employed in this research instrument are valid, as they met the significance criteria and demonstrated sufficiently strong correlations with the total scores of the variables measured.

Table 3. The validation results

Source: Authors

Once the validity of the instrument was confirmed, the next step was to test its reliability using Cronbach’s alpha, a widely applied method for measuring the internal consistency of a scale. Cronbach’s alpha values ranged from 0 to 1, with higher values indicating greater reliability. An instrument can be categorized as moderately reliable when Cronbach’s alpha falls within the range of 0.6-0.7, reliable when between 0.7-0.8, and highly reliable when it exceeds 0.8. In this study, the reliability test yielded a Cronbach’s alpha value of 0.951, indicating that the instrument possessed exceptionally high reliability. This result demonstrates that the research instrument has a very strong internal consistency, meaning that each indicator consistently and stably measures the intended construct. Overall, the findings of this study confirm that the instrument used is both valid and reliable, thereby ensuring that data collection can be conducted with a high degree of confidence and accuracy. These results highlight the importance of conducting validity and reliability tests to guarantee the quality of research instruments, particularly in studies that involve measuring psychological and behavioral dimensions.

The suitability of the dataset for factor analysis was evaluated using the KMO Measure of Sampling Adequacy and Bartlett’s Test of Sphericity. The KMO value obtained was 0.940, indicating excellent sampling adequacy. Bartlett’s Test of Sphericity produced a significant result (χ² = 1306.367; p < 0.05), confirming that the correlation matrix is not an identity matrix (Table 4). These results indicate that the dataset is suitable for factor analysis (factorability), rather than testing data normality. The analysis of the Anti-Image Correlation results within the framework of EFA demonstrates that all indicators in this research instrument meet the requirements for sample adequacy Measure of Sampling Adequacy (MSA). Theoretically, an indicator is considered suitable for inclusion in further analysis if its Anti-Image Correlation value exceeds the threshold of 0.50. This criterion was applied to ensure that each variable made an adequate contribution to the factor structure to be established. Based on the calculations, all MSA values for the respective indicators were above the 0.50 threshold. These findings provide a strong basis for concluding that all indicators are statistically appropriate for further analysis in the factor extraction process. Consequently, none of the indicators needed to be eliminated at this stage, and all variables could be included in the development of valid and reliable constructs.

Table 4. The communalities results

Source: Authors

In the process of EFA, one critical step is the evaluation of the communality’s values for each indicator. The criterion applied to assess the adequacy of indicators at this stage is the extraction communalities value, whereby an indicator is considered acceptable if its extraction value exceeds 0.50. This value represents the proportion of variance in each variable that can be explained by the factors in the model. Based on the results of the calculations, two of the 18 indicators analyzed, I3 and I16, had extraction values below the 0.50 threshold. Accordingly, these two indicators were deemed insufficient to represent the latent construct under investigation and were therefore excluded from the subsequent stages of the analysis. The remaining 16 indicators demonstrated adequate extraction values and were retained for further factor analysis. These findings indicate that the indicators in this study significantly contribute to explaining the factor structure that was established.

The analysis of the variance exploration results indicated that two principal factors emerged from the 16 indicators analyzed. The first factor had an eigenvalue of 8.890 and contributed significantly to explaining 55.561% of the variance in the dataset. This finding suggests that more than half of the information contained within the indicators can be represented by one dominant factor. The second factor had an eigenvalue of 1.115, accounting for an additional 6.967% of the variance. Cumulatively, these two factors explained 62.529% of the total variance, which is considered highly satisfactory in factor analysis, as it exceeds the minimum threshold for model interpretability. The presence of these two factors was further supported by the visual pattern observed in the scree plot, which illustrates the relationship between eigenvalues and the number of extracted factors. The scree point, where the slope of the curve began to level off, was observed immediately after the first two factors (Figure 2). This indicates that these two factors constitute the optimal number within the data structure, whereas subsequent factors contribute negligibly to explaining the additional variance.

Figure 2. Scree plot

The Rotated Component Matrix in this study was used to identify the factor structure of the 16 indicators that had passed the validity and reliability tests (Table 5). The criterion applied in this process was the factor loading value, whereby a variable was considered significant if its loading exceeded 0.55 and was clustered within a specific component. This value reflects the strength of the relationship between an indicator and the factor. Based on the rotation results, two principal components were identified. The first component consisted of 12 indicators with factor loading values above 0.55, all consistently grouped within a single dimension. This indicates that the 12 indicators demonstrated strong correlations and collectively formed a coherent factor, which was subsequently interpreted as Factor 1. The second component comprised the remaining four indicators, each with factor loading values above 0.55, which consistently clustered within Component 2. This consistent grouping suggests that these four indicators represent a construct distinct from that of the first group, thereby forming Factor 2.

Table 5. The rotated component results

Source: Authors

These findings reinforce the structural validity of instrument measurement, as each indicator successfully converged into the factor to which it theoretically belongs (Table 6).

Table 6. The exploratory factor results

Source: Authors

The first factor reflects the structural and institutional dimensions that underpin the overall success of FABA utilization. The indicators included in this factor highlight the importance of adequate infrastructure as a foundational prerequisite to support the distribution and FABA use. Moreover, the level of public understanding on FABA benefit as a construction material directly contributes to its acceptance and adoption at the local level. Community potential for active participation, the technological readiness of small-scale industries, as well as educational programs and information campaigns, reinforce collective awareness FABA usage importance. Additional supporting elements identified include collaborative forums as spaces for innovation, the availability of structured implementation guidelines, the presence of local markets ready to absorb FABA-based products, simplified licensing procedures, and synergy among government, business, and local communities, all of which jointly foster a conducive and sustainable utilization ecosystem. The second factor emphasizes aspects of promotion, regulation, and the strengthening of interactor relationships within the FABA value chain. The indicators grouped under this factor include effective promotion and marketing strategies to expand FABA market acceptance. The commitment of both government authorities and coal-fired power plants to support the implementation of FABA-related regulations serves as a crucial pillar for the continuity of these initiatives. The level of integration between power plants as producers and industrial sectors as end-users represents a crucial factor; when reinforced, it can significantly accelerate the process of utilization. Finally, technical training targeted at MSME actors serves as an important bridge for transferring practical skills and expanding community production capacity in transforming FABA waste into economically valuable products.

Qualitative data analysis in this study employed case-study approach, with the objective of inductively developing themes and concepts derived from field data. The analysis was carried out in three main stages: open coding, axial coding, and selective coding. Semi-structured interviews were conducted with two strategic informants: Pahrinnor, representing Asam-Asam power plant management, and Zainal Arifin, the Village Secretary of Pandansari. The interviews were analyzed using manual coding techniques. Each transcript was carefully examined to extract “core ideas” and “keywords,” which were then developed during the open coding process. The results of open coding were subsequently advanced into the axial coding and selective coding stages to formulate conceptualization, categorization, and thematization of the interview findings (Table 7). Open coding constituted the initial stage of qualitative analysis, conducted by thoroughly reviewing the entire interview transcripts of the two key informants, both of whom held strategic roles in the context of FABA management and utilization. This stage aimed to identify and separate each data segment containing significant meaning into basic units of information. These units of meaning were interpreted as direct representations of the informants’ experiences, perceptions, and perspectives regarding field realities. The process was both descriptive and interpretative, as the researcher systematically organized information into preliminary categories that captured recurring thematic patterns. This analysis produced fourteen major categories, which provided the initial basis for developing broader conceptual categories in the subsequent stage. From the open coding results, units of meaning that shared substantive similarities and thematic tendencies were consolidated during the axial coding stage. This process was performed to construct more meaningful relational patterns among the data units, which were then organized into broader and more structured conceptual categories. Axial coding enabled the researcher to identify relationships among causal conditions, contexts, interventions, strategies, and consequences of the phenomena under study. Through this process, fragmented data was organized into ten thematically significant categories with deeper conceptual meaning. These categories were subsequently employed to understand more complex dynamics in the context of FABA utilization, leading toward data integration in the selective coding stage. Axial coding was then restructured into a broader and more systematic thematic framework that reflected the primary focus of this research. This process resulted in five core elements representing the conceptual framework for understanding critical aspects of FABA management and utilization: economic support, infrastructure, regulation and licensing, socialization, and multi-stakeholder collaboration. Each of these elements serves as an analytical pillar for understanding the dynamics of the system involved. Of the five elements considered, economic support emerged as the most pressing priority, given its direct bearing on the ability of local communities to secure business opportunities, mobilize capital, and sustain operational viability in the utilization of industrial waste. This coding structure offers a robust conceptual framework for advancing the analysis of interrelated factors in ESG context.

Table 7. Manual coding

Source: Authors

Note: FABA = Fly Ash and Bottom Ash

Findings from the qualitative analysis reveal that challenges in FABA utilization arise not only from technological readiness but also from inequities in economic access, infrastructure, and governance. Economic support emerged as the central theme, shaped by systemic issues including the lack of equipment subsidies, training opportunities, and micro-financing schemes (Table 8). These findings underscore that the sustainability of FABA utilization projects is highly dependent on the presence of contextual and well-targeted economic interventions. Furthermore, the interviews confirmed that rural communities demonstrate strong willingness to participate. However, they are constrained by limited access to capital, technology, and market. This reflects a gap between local enthusiasm and the systemic support required. Without adequate interventions, the economic potential of FABA will remain untapped within a fragmented supply chain.

Table 8. Economic support element

Source: Authors

Note: FABA = Fly Ash and Bottom Ash

The findings of this study show that the success of the external utilization of FABA is largely determined by the alignment between infrastructure support, regulatory effectiveness, community engagement, and collaborative governance schemes. The quantitative analysis identified two main constructs, namely 1) economic and infrastructure support and 2) social and regulatory support, while the qualitative analysis revealed five key themes: economic support, infrastructure, regulation and licensing, socialization, and multi-stakeholder collaboration. These findings are explicitly interconnected, where Factor 1 corresponds to economic support and infrastructure readiness, while Factor 2 aligns with socialization, regulatory frameworks, and stakeholder collaboration. This integration demonstrates strong convergence between quantitative factor structures and qualitative thematic insights, reinforcing the robustness of the findings. The divergent perceptions between community members and MSME actors reflect structural asymmetries in knowledge, access, and institutional support. Communities tend to perceive FABA as low-value waste due to limited exposure and lack of demonstration projects, whereas MSMEs recognize its economic potential but face operational barriers such as logistics costs, licensing complexity, and limited technical assistance [2]. These findings suggest that the main constraint is not merely technological, but lies in “last-mile” governance challenges, including weak coordination, insufficient policy incentives, and gaps in knowledge transfer at the local level. Therefore, intervention strategies must be differentiated: community-focused programs should emphasize awareness and social legitimacy, while MSMEs require targeted support in supply chain efficiency, financing, and regulatory facilitation [2, 20].