Methane yield was estimated using empirical relationships between substrate physicochemical properties and methane production reported in previous AD studies [6,10]. Correlation analysis was conducted to evaluate the relationships between methane yield and selected parameters, including pH, TS, VS, and VS/TS ratio. Linear regression analysis was subsequently applied to establish predictive relationships between substrate characteristics and methane yield. The resulting methane yield estimates were converted to theoretical energy recovery potential using the lower heating value of methane.

The AD process can be summarized using the modified Buswell equation. This equation provides a theoretical framework for calculating the amount of methane produced from a given organic substrate [10].

The theoretical methane yield was corrected using biodegradability coefficients (k) obtained from published AD studies involving municipal organic waste. The corrected methane yield was used to estimate realistic methane recovery potential and associated energy production.

A schematic diagram of the two-stage AD system used in this study is presented in Figure 1. The experimental system was designed to evaluate practically achievable methane recovery under continuous AD conditions using BMSW as substrate.

Before digestion, the collected BMSW was shredded and mixed with water at a wet weight ratio of 1:1 to facilitate substrate homogenization and wet anaerobic processing. The prepared slurry was first introduced into a continuously stirred acidification reactor operated under mesophilic conditions (35-37°C) at pH 6.5 for 5 days to promote hydrolysis and acidogenesis. After the acidification stage, insoluble materials were separated using a 1 mm filtration membrane. The resulting liquid fraction was subsequently diluted 1:4 and continuously fed into the methanogenic reactor under mesophilic conditions (35-37°C) at an OLR of 1.3, 2.6, and 5.1 g-VS/L/day. The monitoring period lasted 12 days for each OLR stage. Methane production performance was evaluated after stable reactor operation was achieved.

Biogas generated from the digestion system was collected using gas storage bags and quantified by the volumetric method. Gas composition, including methane concentration, was analyzed using a gas chromatograph (GC-2014, Shimadzu, Japan). The measured methane yield was subsequently used to evaluate experimentally achievable energy recovery and compare practical methane conversion performance with the theoretical and empirical estimation approaches (S1 and S2).

The relationship between selected physical parameters and methane yield is presented in Figure 3. A strong and statistically significant positive correlation was observed between TS and VS (r = 0.92, p < 0.001), indicating that VS represents the biodegradable fraction of TS. However, no significant correlation was found between methane yield and individual parameters such as pH, TS, or VS when considered independently.

In contrast, the VS/TS ratio exhibited a moderate but statistically significant positive correlation with methane yield (r = 0.51, p < 0.05, n = 27), indicating that the quality and biodegradability of organic matter, rather than its absolute quantity, are a more reliable predictor of biogas production. A higher VS/TS ratio generally indicates a greater proportion of biodegradable organic matter available for microbial hydrolysis and methanogenesis, thereby enhancing substrate accessibility and methane conversion efficiency. A linear regression analysis yielded the following relationship: YCH4 = 5.6744 × (VS/TS) - 58.76 (ml/g-VS). Based on this model, a VS/TS ratio of approximately 76% corresponds to a methane yield of 347 mL CH₄/g-VS, equivalent to an energy output of 212 kWh/ton of waste. These values are consistent with reported ranges for organic-rich municipal solid waste systems in Asia [3,26].

Despite its practical usefulness, this approach has inherent limitations. Empirical models do not explicitly account for microbial kinetics, inhibitory compounds, or operational instability, and therefore tend to provide optimistic estimates. Consequently, while S1 is suitable for rapid screening, its predictive accuracy for full-scale systems requires further validation.

Based on the elemental composition presented in Table 3, the biodegradable fraction can be approximated by the molecular formula: C₇₉₇H₁₀₄₅O₁₉₃N₃₉S. The stoichiometric conversion of this substrate yields a theoretical methane potential of 731 mL CH₄/g-VS. However, when applying an average biodegradability coefficient (k = 0.69, see Table 3), the adjusted methane yield decreases to approximately 505 mL CH₄/g-VS, corresponding to an energy recovery of 308 kWh/ton of waste. This estimate is approximately 45% higher than the empirical prediction (S1), consistent with the inherent assumptions of stoichiometric models, which is neglect process inefficiencies and assume near-complete substrate conversion. Similar overestimations have been reported in studies of agricultural and organic waste digestion [10].

It is important to note that the accuracy of this approach depends strongly on the selected biodegradability coefficient (k), which varies with reactor configuration and operating conditions. As shown in Table 3, k values range from 0.61 to 0.83. The use of an average value introduces uncertainty, particularly when extrapolating to specific systems. Therefore, calibration of k under controlled experimental conditions is necessary to improve predictive reliability.

The experimental methane yield obtained in this study reached 334 mL CH₄/g-VS at an OLR of 1.3-5.1 g-VS/L/day, representing the achievable biodegradation performance under controlled AD conditions. This value reflects the combined influence of substrate composition, microbial activity, and operational parameters, and therefore provides a realistic indicator of methane recovery performance for mixed BMSW. When converted into energy terms, the experimental methane yield corresponds to approximately 204 kWh per ton of waste, confirming that even moderate methane production levels can provide a meaningful energy recovery pathway. However, the comparison with higher-performing systems in Figure 4 indicates a clear potential for improvement through process optimization.

A direct comparison with international datasets presented in Figure 4 provides important context for interpreting this result. Methane yields reported across different countries varied considerably, ranging from 282 to 545 mL CH₄/g-VS. Within this distribution, the methane yield obtained in the present study falls in the lower–middle range, closely aligned with values reported for Belgium (317 mL CH₄/g-VS) and slightly higher than those observed in the USA and Spain. These findings suggest that the investigated system performs within the expected range for mixed municipal organic waste digestion systems. However, it does not yet achieve the higher methane recovery efficiencies reported in more optimized AD configurations.

Overall, the experimental methane yield provides a robust and realistic baseline for evaluating AD performance. When interpreted in the context of the global variability illustrated in Figure 4, this highlights both the feasibility of energy recovery from biodegradable municipal waste and the importance of process optimization to achieve higher efficiency.

A comparison among the three evaluation approaches (S1–S2–S3) demonstrates substantial differences between theoretical methane potential and experimentally achievable methane recovery. Among the evaluated methods, S2 generated the highest methane estimates because stoichiometric calculations assume complete substrate conversion under ideal digestion conditions. In contrast, S3 provided lower but more practically representative methane recovery values because experimental systems inherently reflect operational limitations and process inefficiencies.

The observed hierarchy (S2 > S1 > S3) highlights the importance of integrating theoretical and experimental approaches when assessing AD performance for heterogeneous municipal waste systems. While theoretical models are useful for preliminary screening and resource estimation, they may overestimate practical methane recovery if substrate accessibility and operational constraints are not adequately considered. Importantly, the integrated S1–S2–S3 framework provides a more comprehensive evaluation of methane recovery potential by simultaneously accounting for substrate composition, theoretical biodegradability, and experimentally achievable conversion efficiency. Rather than relying on a single estimation approach, this framework enables a more realistic assessment of AD applicability under practical municipal waste management conditions.

These findings further support the conclusion that methane recovery performance depends not only on the quantity of biodegradable waste but also on the physicochemical quality, biodegradability, and process stability. Consequently, optimization of source separation, substrate homogenization, and operational control may be as important as feedstock quantity for improving methane recovery efficiency in municipal organic waste systems.

To contextualize this value, estimates derived from the empirical (S1) and stoichiometric (S2) approaches indicate higher gross energy potentials of approximately 204 kWh/t and 309 kWh/t, respectively. These values define the expected and theoretical upper bounds of system performance. The observed gap of approximately 45-51% between the experimental and theoretical estimates reflects real-world inefficiencies, including incomplete biodegradation, mass-transfer limitations, and operational constraints.

As illustrated in Figure 5, the experimental system achieved lower net energy recovery than the theoretical predictions based on S1 and S2, highlighting the difference between ideal substrate conversion assumptions and practical AD performance under laboratory-scale conditions. The figure further demonstrates that energy recovery potential is strongly influenced by substrate physicochemical characteristics, biodegradability, and process efficiency. These findings support the proposed hypothesis that favorable waste composition can substantially enhance methane recovery and energy-generation potential. However, practical system limitations may reduce achievable performance compared with theoretical expectations.

At a system level, even the conservative estimate of 184 kWh/t represents a meaningful energy recovery pathway for organic waste streams. For large-scale applications, this translates into substantial cumulative energy generation. However, the achievable energy output is highly sensitive to feedstock quality and process optimization. As indicated in Section 3.2, systems operating under optimized conditions such as improved waste segregation, controlled organic loading rates, and advanced reactor configurations can approach the upper range defined by S1 and S2. Therefore, the values presented here should be interpreted as a performance spectrum rather than a single deterministic outcome.

In addition to renewable energy recovery, AD offers significant greenhouse gas mitigation benefits by preventing uncontrolled methane emissions from landfills and partially displacing fossil-based electricity generation. Under conventional landfilling conditions, untreated BMSW is estimated to generate approximately 1.60 ton-CO₂e per ton of waste due to anaerobic decomposition and methane release. This result confirms that landfilling organic-rich municipal waste is a major source of GHG emissions in the development of waste management systems.

The avoided emissions from electricity displacement vary with the efficiency of methane recovery. The empirical model (S1) yields the highest electricity-related mitigation potential (0.115 ton-CO₂e/ton-waste), followed by the equilibrium model (S2) and the experimental model (S3), which generate 0.079 and 0.076-CO₂e/ton-waste, respectively. The differences reflect the variations in methane yield and the corresponding electricity recovery capacity among the three estimation methods.

Methane leakage from the AD system remains relatively low due to the assumed high collection and energy conversion efficiencies. The estimated fugitive emissions range from 0.035 to 0.053 ton-CO₂e/ton-waste. Consequently, the overall net GHG reduction potential remains consistently high across all evaluated scenarios, ranging from approximately 1.55-1.57 ton-CO₂e per ton of treated waste.

These findings indicate that the environmental benefits of AD are predominantly driven by avoiding landfill methane emissions. At the same time, electricity displacement provides a complementary but relatively smaller contribution to the total GHG mitigation. Although the differences among the methane recovery scenarios S1–S3 moderately affect energy-related emissions, the overall GHG reduction potential remains relatively stable because the avoidance of landfill methane dominates the system-level carbon balance. Overall, the results demonstrate that AD represents a promising low-carbon strategy for managing organic-rich municipal waste streams, particularly in regions where landfilling remains the predominant disposal pathway.

Capital expenditure remains a major barrier to implementing AD infrastructure, particularly in developing countries. Reported international datasets indicate that capital investment for AD facilities generally ranges from approximately 110-535 USD/t of annual treatment capacity in Asian countries and 350-500 USD/t in Europe and North America [20]. Based on the compiled database used in this study, the average capital investment was approximately 400 USD/t of treatment capacity, with a strong relationship observed between facility scale and total investment cost (R² = 0.78). As shown in Figure 6, specific capital costs decreased from approximately 373 USD/t for small-scale facilities (100 t/day) to approximately 274 USD/t for large-scale systems (1000 t/day), demonstrating significant economies of scale. Similar trends have been reported in previous techno-economic analyses of organic waste AD systems.

Operational expenditure is another important factor governing AD feasibility. Previous studies reported that operating costs for AD facilities in Asia typically range from approximately 20-40 USD/t of waste processed, while systems in Europe and North America generally exhibit higher operational costs of approximately 50-90 USD/t [19]. These differences are associated with labor costs, maintenance requirements, technological complexity, and process automation levels.

Based on the baseline net recoverable electricity of approximately 184 kWh/t derived from S3, the potential electricity revenue was estimated at approximately 16 USD/t of waste using the maximum biomass electricity purchase price of 2,091.74 VND/kWh (approximately 0.082 USD/kWh) established under Decision No. 1008/QĐ-BCT issued by the Vietnamese Ministry of Industry and Trade in 2025. This Vietnamese case is presented as an illustrative example within the ASEAN context. Under these conditions, electricity recovered alone could offset approximately 35–70% of the reported operating costs for AD systems in Asia. Nevertheless, the economic benefits of AD systems extend beyond electricity generation. In practical municipal waste management systems, AD facilities may also receive tipping fees for processing biodegradable waste, thereby providing an additional and relatively stable revenue stream independent of electricity market fluctuations. Previous studies have further highlighted that renewable energy incentives, feed-in tariffs, and carbon credit mechanisms substantially improve the profitability and implementation potential of AD systems [9,16].

In addition to electricity generation, the economic viability of AD systems may be further enhanced through digestate valorization and nutrient recovery strategies within integrated circular-economy frameworks. Nevertheless, several uncertainties remain when extrapolating laboratory-scale performance to full-scale implementation. Variability in waste composition, insufficient source separation, and operational instability may reduce methane recovery efficiency and increase treatment costs under practical conditions. Therefore, although the present findings indicate that AD represents a technically feasible and potentially economically attractive strategy for managing organic-rich municipal waste, further pilot-scale and long-term techno-economic validation remains necessary to improve the reliability of full-scale implementation assessments.