Optimizing Feedstock Selection for Sustainable Small-Scale Biogas Systems Using the Analytic Hierarchy Process ^†

Abstract

Small-scale biogas systems can play a pivotal role in sustainable energy provision, particularly in developing countries. However, their dependence on livestock manure as the only feedstock poses challenges to their adoption and long-term viability. This often leads to insufficient biogas production and plant abandonment. This study proposes co-digestion of livestock manure with other farm residues to enhance the technical sustainability of small-scale biogas systems by ensuring adequate and consistent biogas production throughout the plant’s lifespan, minimizing the risks associated with reliance on a single feedstock. A novel feedstock selection approach is developed using the Analytic Hierarchy Process (AHP), a multicriteria decision-making method, to prioritize feedstocks based on adequacy, supply consistency, and logistical ease. AHP is chosen due to its capability to handle both quantitative and qualitative evaluation criteria. This approach is applied to the Fès-Meknès region of Morocco, which offers abundant livestock and crop residues alongside product utilization pathways. The prioritization and ranking of the potential feedstocks identified in the region reveals cattle manure as the top-ranked feedstock due to its consistent supply and ease of collection, followed by straw, valued for its storability and nutrient stability. Sheep, horse, and chicken manure ranked third, fourth, and fifth, respectively, while household food waste and fruit and vegetable residues, limited by seasonality and perishability, were ranked lower. Based on these findings, co-digestion of cattle manure and straw is proposed as a sustainable strategy for small-scale biogas plants in Fès-Meknès, addressing feedstock shortages, enhancing biogas production, and reducing plant abandonment. This approach strengthens technical sustainability and promotes the broader adoption of biogas technologies in developing countries.

1. Introduction

1.1. Background and Context

Feedstock selection is a crucial step when setting up a biogas plant, whether for small-scale household use or large-scale combined heat and power (CHP) or biomethane production. It impacts the plant design and biogas production technology employed, preprocessing steps, operational costs, project viability, and sustainability. Incorrect feedstock selection can make biogas production unreliable, expensive, and complex, rendering it unsustainable. However, feedstock selection is not straightforward; various, often conflicting factors must be considered. Owing to their size, level of technology, and the economic ability of small-scale biogas systems users and adopters in developing countries, feedstock selection is even more significant. Given these challenges, this study develops a practical approach that enables biogas technology stakeholders to account for these factors in feedstock selection, minimizing the risks associated with reliance on a single feedstock.

Small-scale biogas systems, also called ‘domestic’, or ‘household’ biogas systems, are typically operated primarily on livestock manure as feedstock. A fluctuation of livestock numbers due to drought, slaughter, or reproduction cycles, among others, affect the continuous functioning of digesters. Energy demand for the household will therefore need to be met either by other sources or crop/vegetable residue will need to be fed into the digester [1]. The resilience and continuous operation of these digesters can only be assured by careful planning and selection of feedstocks. Identification and quantification of feedstocks is the beginning to ensuring sustained biogas production [2]. Farms are an ideal starting point for rolling out small-scale biogas plants in developing countries, as they provide a unique combination of feedstocks, energy utilization opportunities, and digestate application options, which are central considerations for establishing a biogas plant.

Biogas produced from small-scale biogas plants plays an important role in providing clean energy for cooking and heating in developing countries, replacing high-emission fuels like charcoal, and firewood [3]. This transition is vital for reducing indoor air pollution, which was responsible for over 2.3 million deaths and the loss of more than 91 million disability-adjusted life years in 2019 [4]. Additionally, biogas can be used to power gas engine-driven water pumps for irrigation and household purposes.

In 2020, approximately 38 billion cubic meters of biogas were produced globally, equivalent to 1.46 exajoules of energy. Of this production, Europe accounted for 53%, Asia contributed 32%, and the Americas contributed 13% [5]. This contrasts with IEA reports indicating that developing countries currently account for two-thirds of the about 600 Mtoe of sustainable global biogas potential. Specifically, developing countries in Asia have 30%, those in Central and South America have 20% of this global biogas potential, and Africa, although having a smaller potential, possesses enough potential to meet the needs of the 600 million people without electricity access in sub-Saharan Africa [6]. Nevertheless, the dissemination and adoption of small-scale biogas systems has been ongoing since the 1970s in developing countries, with varying levels of success [7]. For instance, China and India have seen significant adoption, with a total of over 50 million units installed [3,8], including the introduction of medium- and large-scale plants. However, many nations in the Global South and Asia are yet to establish biogas as a mainstream energy source despite numerous governmental and non-governmental initiatives.

Small-scale ‘household’ biogas plant installation and adoption in developing countries is currently conducted based on access to feedstock, mainly manure from livestock (at least three heads of cattle, at least ten pigs in zero-grazing, or a daily access to at least 5 kg of kitchen waste), and water, among others [9]. However, the reliance on livestock manure as the basis for rollout of household biogas plants has been plagued by numerous challenges, as seen in the following cases: (a) complete abandonment due to inability to sustain livestock keeping and (b) inadequate feeding of the digesters as a result of insufficient manure production:

• Studies in Uganda by Lwiza et al. [10] found that biogas plants were often abandoned within four years due to several factors including drought (leading to inability to keep enough livestock), disease, shifts in land use priorities, and technical issues with digester performance. Similar findings were reported by Paramonova et al. [11] in Vietnam, where these challenges led to dis-adoption of small-scale biogas systems. In China, pig manure is the primary feedstock for household biogas digesters. However, in 2018, an outbreak of African Swine Fever (ASF) led to the culling of over 1.2 million pigs, resulting in a severe manure shortage and the eventual shutdown of many biogas digesters due to unsustainable feedstock supply [12]. Bhat et al. [13] noted that reductions in manure availability and an increase in alternative uses often lead to insufficient biogas production, causing households to abandon biogas systems and revert to traditional, polluting cooking methods.
• In Bangladesh, Rahman et al. [14] observed that cattle and chicken farmers underfed their digesters with about one-third less manure than recommended. A review of Rwanda’s National Domestic Biogas Program (NDBP) by FAO [15], revealed that over 20% of biodigester operators had fewer cows than needed to produce sufficient manure to sustain biogas production, while another one-quarter had their digesters producing less biogas than expected.

These challenges and failures highlight the weaknesses of biogas dissemination programs that rely solely on livestock manure. It is therefore crucial to identify alternative complementary feedstocks that can be utilized in small-scale biodigesters to enhance their sustainability, resilience, and long-term operation [7,15].

The failure and abandonment of biogas digesters demoralizes potential adopters and damages the technology’s reputation, creating obstacles to its widespread adoption among farmers in developing countries [12,16]. This underscores the dual objectives of this study. First, by developing a novel systematic approach for prioritizing, ranking, and selecting optimal and sustainable feedstocks and feedstock combinations for small-scale biogas plants, operators and adopters can better anticipate potential reductions in availability of livestock manure. This approach aims to prevent underfeeding and, ultimately, the abandonment of biogas plants. Second, by applying this approach in a case study area, where a survey identified potential feedstocks (showing that both livestock manure and crop residues are prevalent), its applicability is demonstrated, and a framework is provided for regions where the technology is still emerging.

This proactive strategy seeks to address the challenges associated with traditional methods of deploying and operating small-scale biogas plants while also offering a pathway to potentially redesigning existing plants. By enhancing the resilience and sustainability of small-scale biogas production, this approach aims to promote the broader adoption of biogas technologies as a vital renewable energy source for households and farms in developing countries.

1.2. State-of-the-Art

Renewable energy systems, such as anaerobic digestion for biogas production, wind, and solar, are typically evaluated based on four key factors: financial/economic, technical, social, and environmental. However, in biogas production, the critical first step is feedstock selection, which underpins all subsequent analyses [17]. The installation and adoption of small-scale household biogas plants in developing countries often depend on access to key feedstocks, primarily livestock manure (from cattle or pigs) and water, among other factors [9]. While decision-makers often apply a triple-bottom-line approach, considering environmental, social, and economic factors to assess the sustainability and viability of renewable energy systems, this study explores the feedstock selection process, which significantly impacts the technology, design, and long-term operation of biogas plants.

Decision-making in the renewable energy sector has often utilized methods like the weighted sum method (WSM) and the weighted product method. Jovanović et al. [18] applied the weighted sum method to evaluate sustainable development scenarios for Belgrade’s energy system and to assess and select suitable energy systems in Bosnia. Due to its simplicity, WSM forms the foundation for complex aggregation methods such as analytical hierarchy process (AHP) [19].

The Analytic Hierarchy Process, developed by Saaty in the 1970s [20], is a widely used multi-criteria decision-making (MCDM) method. It leverages pairwise comparisons to evaluate different criteria and alternatives, effectively handling both tangible and intangible factors in decision-making [19]. AHP has been extensively applied in renewable energy systems for technology selection, site optimization, and improving decision-making processes by addressing the complexity of multiple, often conflicting, criteria. For example, AHP has been used to prioritize renewable energy generation and assess floating photovoltaics in Laos [21]. In Pakistan, Mirjat et al. [22] employed AHP for a multi-criteria assessment of electricity generation scenarios, considering technological, environmental, socio-political, and economic factors for sustainable energy planning.

Moreover, Ossei-Bremang and Kemausuor [23] developed a decision support system using a hybrid fuzzy technique for order preference by similarity to ideal solution (FTOPSIS) and AHP to select sustainable biomass resources for energy production in Ghana. By applying sustainability criteria (economic, social, and environmental) to various biomass resources (including animal manure, crop residue, energy crops, municipal solid waste, and forest residue), they identified the most suitable biomass resources for Ghana.

MCDM methods, particularly AHP, have also been used for renewable energy site selection, as reviewed by Shao et al. [24]. Many studies they reviewed applied AHP’s pairwise comparison method to determine weights due to its simplicity, understandability, and robust outcomes. On the biogas technology front, AHP was applied to rank barriers to the adoption of small-scale biogas plants in rural India [25] and in the prioritization and selection of the most appropriate process of anaerobic digestion for energy production from biomass in Iran [26].

In addition to AHP and AHP-based methods, other multicriteria decision-making (MCDM) methods applied in the renewable energy sector include a family of methods called ELECTRE (ELimination Et Choix Traduisant la REalité). ELECTRE, a non-compensatory approach developed by Roy [27], is based on the principle that the decision-maker is not perfectly rational and can express varying levels of preference between alternatives, including strong preference, indifference, weak preference, or non-comparability. ELECTRE is particularly useful when trade-offs between criteria are not desirable or feasible. An alternative to ELECTRE is the Technique for Order Preference by Similarity to Ideal Solutions (TOPSIS). This method selects the best solution by minimizing the distance to an ideal solution while maximizing the distance from a negative ideal solution. While effective, this method does not account for the relative importance of criteria when determining the ranking index and distances from the ideal and negative ideal solutions. Another widely used method is the Preference Ranking Organization Method for Enrichment Evaluation (PROMETHEE). PROMETHEE provides a simple yet effective way to rank alternatives based on pairwise comparisons and preference functions, making it particularly useful for complex decision problems with multiple conflicting criteria [28].

In this study, AHP was chosen due to its strengths outlined below [22,29]:

• Objective and subjective evaluation: It assesses both objective and subjective functions in multicriteria decision-making, aiding in reaching a consensus.
• Optimal judgment guidance: It directs decision-makers towards the optimal judgment for the issue at hand rather than seeking a definitive ‘correct’ solution.
• Hierarchical structure: It provides a broad and balanced framework for decision-making, organizing large problems into smaller, manageable subproblems.
• Criteria prioritization: It identifies and prioritizes significant factors by weighting different criteria, clearly indicating their relative importance numerically.
• Consistency check: It includes the calculation of the consistency ratio, a key feature of AHP, allowing verification of the consistency and rationality of judgments, thus reducing bias.
• Versatility: Applicable to both qualitative (intangible) and quantitative (tangible) criteria, making it adaptable to a wide range of decision-making scenarios.
• Ease of use: Simple to understand and apply, even for complex issues.

To date, studies or guidelines for developing small-scale biogas programs that utilize multicriteria decision-making methods in feedstock selection could not be identified.

1.3. Novelty of the Study

The novelty of this study lies in developing a practical approach that guide biogas plant adopters and operators in optimizing feedstock selection, considering multiple criteria that directly influence plant design, bio-digestion and pretreatment technologies, operational costs, and long-term sustainability. By providing a structured and better-informed data-driven method for feedstock selection, this research enhances the planning, design, and operation of small-scale biogas plants, ensuring more reliable and efficient energy production.

This paper is structured as follows:

Section 2: Materials and Methods—Systematic Approach to Feedstock Selection: This section begins with an overview of the methodology used in this research. The alternatives to be considered are then presented, followed by a literature review to identify suitable feedstock selection criteria and sub-criteria. Finally, the application of the AHP methodology to feedstock ranking and prioritization is outlined, showing clearly the hierarchical structure of the process.

Section 3: Results and Discussion: This section presents and discusses the feedstock ranking and prioritization results. Applying the AHP methodology, the feedstocks that could potentially improve the resilience and sustainability of biogas plants in the case study area are ranked and selected.

Section 4: Conclusion and recommendations: This section provides a summary of the main research findings, addresses the limitations of the study, and outlines recommendations for future research.

2. Materials and Methods

This section outlines the methodology for developing a systematic process to prioritize and select biogas plant feedstocks for small-scale biogas plant researchers, adopters, and implementers. The approach outlines criteria, sub-criteria, and steps for selecting feedstocks and their possible combinations to enhance the resilient and sustainable operation of small-scale biogas plants. First, a survey was conducted in the case study area of Fès-Meknès in Morocco to identify potential biogas plant feedstocks as described in [30]. Secondly, the selection criteria and sub-criteria were identified from the literature. Then, AHP was used to develop the selection process, which was then applied to the feedstock alternatives identified in the case study area. The methodology applied in this study is summarized in Figure 1.

2.1. Feedstock Alternatives in the Fès-Meknès Region

The goal of the decision-making process was to rank feedstocks that would enhance sustainability of biogas production in small-scale biogas plants. The systematic approach was applied to Morocco’s Fès-Meknès region where a survey had been conducted to identify potential biodigester feedstocks for farm-based biogas plants. The Fès-Meknès region was chosen as the case study area due to its rich biomass resources from the agricultural and forestry sectors boasting about a fifth of the country’s total potential. This region’s useful agricultural area makes up about 15% of the national total. Additionally, the region supports a significant livestock population, including around 3 million sheep, 426,000 cattle, and 424,000 goats [31,32]. Figure 2 provides an illustration of the case study area.

A survey utilizing a stratified sampling methodology as described in [30] was conducted to identify potential farm-based biogas plant feedstocks in the form of livestock and crop wastes. The survey also gathered data on current residue utilization, household sizes, willingness to adopt biogas technology, and other relevant factors among local farmers in the Fès-Meknès region. The results obtained showed that most farms in the region rear, on average, 11 cattle, 45 sheep, 9 goats, and 20 chicken. A few farms also have an average of 2 horses [30]. In addition, all the farms with livestock indicated that they keep their animals in a shed or enclosure for a part or the whole of the day, facilitating the collection of a significant portion of produced manure.

Table 1. Average livestock numbers and potential feedstock amounts in the case study area of Fès-Meknès [30].

Livestock Type	Number of Farms (N)	Mean	Potential Feedstock (kg/day/head) *	Average Potential Feedstock (tonnes/year/farm)
Cattle	44	11 ± 19	10	40.15
Sheep	39	45 ± 59	1	16.43
Goats	6	9 ± 5	1	3.29
Chicken	33	20 ± 12	0.08	0.58
Horses	16	2 ± 2	10	7.30

* Adapted from [33].

summarizes the livestock numbers and average potential feedstock amounts per farm from livestock in the case study area.

Additionally, four out of five farms engage in both crop farming and livestock rearing, cultivating cereals (wheat, barley, and maize), vegetables (potatoes and onions), and fruits. The prevalence of crop farming is as shown in

Table 2. Crops grown in the Fès-Meknès region.

Crop Type	Number of Farms
Wheat	33
Barley	16
Onions	16
Potatoes	15
Fruits	12
Maize/corn	8
Other crops *	15

Source: Adapted from [30]. * Other crops include olives, forage, almonds, chickpeas, and walnuts.

.

Farms growing cereals make up over two-thirds of the farms, representing both small-scale subsistence farms and large-scale farms. Cereals are the leading crop harvested and consequently produce the most crop residue in the region [30], reflecting a similar trend with Moroccan government statistics showing that cereals (wheat and barley) represent the largest crop yields in the country and that their farming occupies about 59% of the useful agricultural area [34,35]. A summary of annual cereal yields as reported by farmers surveyed in Fès-Meknès is as shown in

Table 3. Annual cereal crop yields in Fès-Meknès.

Yield Range * (Tonnes)	Number of Farms
	Wheat	Barley	Maize
<1	8	4	4
1 to 5	15	4	2
6 to 10	4	1	1
11 to 20	1	1	1
21 to 30	1	0	0
Over 30	1	1	0

Source: Adapted from [30]. * Most farmers indicated an estimated yield, especially those with small farms.

.

Farms rearing livestock and engaging in crop farming are known producers of biogas plant feedstocks and provide ample avenues for utilization of produced biogas and by-product digestate. These findings agree with IEA [6], where they state that in developing countries, particularly where the agricultural sector is a key economic pillar, crop residues and livestock manure constitute the largest fraction of biogas plant feedstocks.

These residues from livestock and crop farming in the case study area are primarily used to make compost manure, with some being disposed of or burned in the fields while still raw. Only about 15% of farms use crop residues as livestock feed. The residues used for manure and those disposed or burned can be utilized for biogas production, generating valuable energy without compromising organic fertilizer production or soil health since digestate is a concentrated and ready-to-use organic fertilizer and can be applied to the crop fields [30].

From these results, therefore, fruit and vegetable waste (potatoes, onions, and fruits), manure from cattle, sheep, chicken, and horses, and straw (from wheat, barley, and corn) are considered as potential biogas plant feedstocks in this region. These agricultural residues comprise the alternatives to be considered in this study alongside household food waste. It is estimated that household food waste is produced at a rate of 75–100 kg/year per capita for all nations, irrespective of the level of development [36]. With an average household size of about $7$ members, as shown by the survey, each household in Fès-Meknès produces, on average, up to 0.7 tonnes of food waste per annum. Food waste is obtained from the food preparation process in form of inedible parts and leftover food from households. It is therefore considered a potential feedstock alongside the agricultural residues.

2.2. Feedstock Selection Criteria and Sub-Criteria

The feedstock selection criteria and sub-criteria for prioritizing the potential feedstock alternatives are derived from the literature. Roubík and Mazancová [37] and Liebetrau et al. [38] emphasize that the successful implementation and sustainability of small-scale biogas technologies depend on key factors, including a consistent, year-round feedstock supply, household purchasing power, efficient biogas plant design, access to materials and skilled labour, and effective utilization of the produced energy and digestate. In addition, important considerations for feedstock supply are the type of feedstock, its convenient daily availability, and minimal labour requirements for handling both feedstock and digestate

Abubakar [39] and Singh et al. [40] highlight critical considerations for biogas plant feedstocks, including feedstock type, supply logistics, sustainability, availability, quantity, nutrient content, and supply chain management. Similarly, Ossei-Bremang and Kemausuor [23] emphasize that biomass collection and storage, particularly for seasonal crops, significantly impact the overall cost and efficiency of the value chain. Moreover, the limited availability of dung (due to a small number of cattle) and its competing uses have often resulted in the swift abandonment of biogas plants, as they fail to provide adequate biogas to meet the cooking energy needs of users [13]. These considerations collectively form the basis for the three main criteria and eight sub-criteria shown in

Table 4. Criteria and sub-criteria for feedstock prioritization and selection.

Criteria	Sub-Criteria	Description	Reference
C1: Adequacy	SC1: Quantity	This criterion evaluates whether the potential feedstock amount is available in sufficient quantities for use as feed to a biogas plant.	[38,39]
	SC2: Suitability	This criterion analyses whether the potential feedstock has sufficient nutrient quantities and whether it is readily digestible in its collected form in a biogas plant.	[39,40]
C2: Supply Consistency	SC3: Seasonality	This criterion looks at whether there are seasonal fluctuations in feedstock supply and availability.	[37,38]
	SC4: Alternative uses	This criterion evaluates whether a particular feedstock typically has other competing uses that may affect its availability for use in biogas production. Feedstocks that have popular alternative uses are most often unavailable for utilization in biogas production.	[13]
	SC5: Storability	Storability is evaluated by considering whether the feedstock can be stored for extended periods without significant nutrient loss or degradation. This is particularly important for seasonal feedstocks. Storability minimizes the impacts of seasonal (long-term) and sudden shortages.	[23]
C3: Logistical ease	SC6: Collection and handling	This criterion analyses the supply chain processes that are involved in the collection, handling, and transportation of the feedstock. It looks at whether the feedstock is typically collected within the vicinity of the biogas plant.	[23,37,40]
	SC7: Operational storage	This criterion considers whether the feedstock requires storage to ease operations and act as short-term storage buffers.	[23,40]
	SC8: Preprocessing	This criterion evaluates whether the feedstock requires some sort of preprocessing after collection, hence requiring some sort of technology or equipment and, consequently, maintenance.

applied in this study for feedstock prioritization, ranking, and selection.

2.3. Application of AHP to Feedstock Prioritization and Selection

The structure of the AHP used in this study consists of four main levels, as shown in Figure 3. The first level represents the goal of the decision-maker, the second level is the criteria, the third level is the sub-criteria, and the fourth level contains the alternatives. Each level is assessed based on the level immediately above it.

To implement the AHP algorithm, the following steps, summarized in Figure 4, are followed [21,22,42]:

• Determine the goal, criteria, and alternatives;
• Using pairwise comparisons, create a set of judgements for all criteria;
• Calculate the relative weights of the different criteria;
• Verify the consistency of the judgements;
• Apply the same method to prioritize the different sub-criteria;
• Repeat for the different feedstocks with respect to each sub-criterion;
• Rank the different feedstocks based on the relative weights of the criteria, sub-criteria, and the prioritization with respect to each sub-criterion.

2.3.1. Pairwise Comparison Matrices

To apply the AHP method to feedstock selection, pairwise comparison matrices for the criteria, sub-criteria, and potential feedstocks need to be generated. Since there is no explicit scale for assigning weights to the criteria, the perceived importance of each criterion to the overall objective of sustainable biogas plant operation is used to determine the criteria weight matrix. The same judgment is applied to the sub-criteria with respect to the different associated criteria. The scale provided in

Table 5. Typical pairwise comparison scale for the AHP method (Modified from [42]).

Intensity of Importance, ${\mathit{a}}_{\mathit{i}\mathit{j}}$	Explanation
1	Criterion/alternative i and j are equally important
3	Moderate importance of i-th criterion/alternative over the j-th
5	Strong preference of i-th criterion/alternative over the j-th
7	The i-th criterion/alternative is strongly favoured over the j-th, and its dominance is demonstrated in practice
9	The i-th criterion/alternative is absolutely favoured over the j-th
2, 4, 6, 8	Intermediate values between two adjacent judgements

is used to develop the pairwise comparison matrices for the different criteria, sub-criteria, and alternatives.

The pairwise comparison matrix is a square matrix of the form shown in Equation (1), where the elements $a_{ij}$ are the relative importance/judgements of criterion/alternative $i$ with respect to criterion/alternative $j$. These pairwise matrices require $n(n-1)$ judgements to be made, where $n$ is the matrix size (that is, the number of elements being compared).

$$A=\left[a_{ij}\right]=\left[\begin{array}{ccccc}1& a_{12}& .& .& a_{1n}\\ {\displaystyle \frac{1}{a_{12}}}& 1& .& .& a_{2n}\\ .& .& .& .& .\\ .& .& .& .& .\\ {\displaystyle \frac{1}{a_{1n}}}& {\displaystyle \frac{1}{a_{2n}}}& .& .& 1\end{array}\right]$$

(1)

2.3.2. Consistency of the Judgements

To test the consistency of the pairwise comparison matrices, the consistency ratio is evaluated. Inconsistency and/or uncertainty occurs due to the vagueness of human judgment [43]. The existence of consistency is not necessarily an indication of accuracy of the choices but rather reflects that there exists rationality in the judgements of the decision maker. For instance, if $a_{ij}$ is the importance of alternative $i$ over $j$ and $a_{jk}$ is the importance of alternative $j$ over $k$, then it follows that for consistency in the judgements, the importance of alternative $i$ over $k$ must be $a_{ij}{a}_{jk}$. To calculate the consistency ratio, the consistency index and random consistency index are used [41,42].

The consistency ratio (CR) is defined as follows:

$$CR={\displaystyle \frac{CI}{RI}}$$

(2)

where:

Consistency index (CI)

$$CI={\displaystyle \frac{{\lambda }_{max}-n}{n-1}}$$

(3)

${\lambda }_{max}$ is the eigenvalue of the corresponding pairwise comparison matrix and $n$ is the number of elements being compared. The eigenvalue is estimated as follows [42].

$${\lambda }_{max}{w}_i=\sum _{j=1}^n{a}_{ij}{w}_j$$

(4)

$w_i$ and $w_j$ are the weight of importance of the $i$-th and $j$-th criterion, respectively.

$RI$ is a random consistency index; it is dependent on the size of the matrix, as shown in

Table 6. Random consistency index [42].

Size of Matrix (n)	1	2	3	4	5	6	7	8	9	10
Random Consistency Index (RI)	0	0	0.58	0.9	1.12	1.24	1.32	1.41	1.45	1.49

. Generally, a $CR$ value less than 0.1 is acceptable. If the CR is higher than 0.1, the pairwise comparisons should be reviewed and revised to improve coherence.

The resultant pairwise comparison matrices, after being reviewed for consistency, are then normalized to prioritize the criteria. To generate the standardized matrix, each element of the pairwise comparison matrix is divided by the sum of all elements in the same column. The elements of the standardized matrix are then averaged row-wise to determine the weights of each criterion. The same is done for the alternatives with respect to the sub-criteria. The completed decision matrix takes the form shown in Equation (5) for $m$ alternatives and $n$ criteria [42]:

$$\begin{gathered} \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\begin{array}{cccc}{c}_1& .& .& c_n\end{array} \\ D=\left[\begin{array}{c}{a}_1\\ .\\ a_m\end{array}\right]\left[\begin{array}{cccc}{v}_{11}& .& .& v_{1n}\\ .& .& .& .\\ v_{m1}& .& .& v_{mn}\end{array}\right] \end{gathered}$$

(5)

where $c_n$ is the criteria or sub-criteria weights and $v_{mn}$ is the weight of alternative $m$ with respect to criterion/sub-criterion $n$.

To rank the different feedstocks, their overall weights are determined by performing the operation given in Equation (6) [42]. The resulting overall weights of each alternative are then used to rank, screen, or choose an alternative or a combination of alternatives. The most suitable alternative or alternatives is the one with the highest score.

$$R=\left[\begin{array}{c}{a}_1\\ .\\ a_m\end{array}\right]=\left[\begin{array}{cccc}{v}_{11}& .& .& v_{1n}\\ .& .& .& .\\ v_{m1}& .& .& v_{mn}\end{array}\right]\times \left[\begin{array}{c}\begin{array}{c}{c}_1\\ .\\ .\end{array}\\ c_n\end{array}\right]$$

(6)

3. Results and Discussion

3.1. Determination of Criteria Weights

The weights assigned to each criterion are determined based on their importance to the sustainable continued operation and optimal performance of small-scale biogas systems. The highest weight is given to the criterion that, when not met, poses the greatest risk of leading to biogas plant abandonment. Based on these considerations, the pairwise comparison matrix shown in

Table 7. Criteria pairwise comparison and normalized matrices.

Pairwise Comparison Matrix				Normalized Matrix
	C1	C2	C3		C1	C2	C3	Weight
C1	1.00	0.50	3.00	C1	0.30	0.29	0.33	0.309
C2	2.00	1.00	5.00	C2	0.67	0.59	0.56	0.581
C3	0.33	0.20	1.00	C3	0.10	0.12	0.17	0.110
Total	3.33	1.70	9.00

${\lambda }_{max}=3.0037, CI=0.0018, CR=0.0032.$

is developed. To generate the standardized matrix, each element of the pairwise comparison matrix is divided by the sum of all elements in the same column. The elements of the standardized matrix are then averaged row-wise to determine the weights of each criterion.

The weighting of the criteria shows that the criterion C2, supply consistency, has the highest impact on the sustainability and long-term operation of small-scale biogas plants. This observation aligns with the literature, which indicates that the inability to consistently provide sufficient feedstock to biogas plants in developing countries, due to factors such as drought and disease, has led to the abandonment of this technology. Criterion C1, adequacy, follows, as both the quantity of feedstock available and its suitability for biogas production significantly affects the continuous economical operation of the biogas plant. Criterion C3, logistical ease, is ranked last, as most feedstock for small-scale biogas plants is sourced from within the farm and is typically gathered during regular farming activities. However, some feedstocks must be collected and transported from their production site to the biogas plant, introducing some logistical considerations.

3.2. Determination of Sub-Criteria Weights

The relative sub-criteria weights are determined from their pairwise comparisons, given in Appendix A, within their respective main criteria as shown in

Table 8. Sub-criteria weights and consistency ratios.

Level 2: Criteria	Level 3: Sub-Criteria	Sub-Criteria Weights	Consistency
C1: Adequacy	SC1: Quantity	0.500	$CR=NA$
	SC2: Suitability	0.500
C2: Supply Consistency	SC3: Seasonality	0.260	${\lambda }_{max}=3.0387$ $CI=0.0194$ $CR=0.0334$
	SC4: Alternative uses	0.106
	SC5: Storability	0.633
C3: Logistical ease	SC6: Collection and handling	0.669	${\lambda }_{max}=3.0070$ $CI=0.0035$ $CR=0.006$1
	SC7: Operational storage	0.088
	SC8: Preprocessing	0.243

. For criterion C1 (feedstock adequacy), both sub-criteria SC1 (quantity) and SC2 (suitability) are considered to contribute equally. When evaluating criterion C2 (supply consistency), sub-criterion SC5 (storability) has the greatest effect at 0.633, followed by SC3 (seasonality) at 0.260, and SC4 (alternative uses) at 0.106. This indicates that a feedstock that can be stored for long periods is more likely to be consistently available for biogas production, even if it may be seasonal or has competing alternative uses. Good storability provides a buffer and allows planning for shortages caused by seasonal variations, sudden unexpected events, or increased popularity of alternative uses.

In terms of criterion C3 (logistical ease), sub-criterion SC6 (collection and handling) has the greatest impact at 0.669, followed by SC8 (preprocessing) at 0.243, and SC7 (operational storage) at only 0.088. Collection and transportation of feedstocks account for most of the delivered cost, particularly for those procured from outside the biogas plant site [23]. On the other hand, operational storage is a one-time fixed cost and thus has a minimal impact on the overall logistical ease of the feedstock supply.

A standardization of the sub-criteria weights is conducted by dividing each individual sub-criterion weight with the sum of all the sub-criteria weights in Table 8 to give their relative weights and ranking with respect to the other sub-criteria as shown in

Table 9. Relative weights of the evaluation criteria and sub-criteria.

Level 2: Criteria	Criteria Weight	Rank	Level 3: Sub-Criteria	Standardized Sub-Criteria wt	Rank
C1: Adequacy	0.3092	2	SC1: Quantity	0.1667	3
			SC2: Suitability	0.1667	3
C2: Supply Consistency	0.5813	1	SC3: Seasonality	0.0868	5
			SC4: Alternative uses	0.0354	7
			SC5: Storability	0.2111	2
C3: Logistical ease	0.1096	3	SC6: Collection and handling	0.2229	1
			SC7: Operational storage	0.0294	8
			SC8: Preprocessing	0.0810	6

.

These results indicate that collection and handling effort, ability to store feedstock, the available quantities, and their ease of digestibility, followed by seasonal variations, are the most critical factors influencing the selection of biogas plant feedstocks in the case study area.

3.3. Prioritization of Potential Biogas Plant Feedstocks

The feedstocks identified in the case study area were assessed using the AHP MCDM method. The pairwise comparisons of the feedstocks with respect to the sub-criteria were used to come up with the prioritization of feedstocks with respect to each criterion by computing the normalized matrices and weights as shown in Appendix B. The weighted feedstock prioritization per sub-criterion was then used to create the decision matrix in

Table 10. Feedstock prioritization with respect to sub-criteria.

		Sub-Criteria
		SC1	SC2	SC3	SC4	SC5	SC6	SC7	SC8
		16.67%	16.67%	8.68%	3.54%	21.11%	22.29%	2.94%	8.10%
A1	Cattle manure	0.4102	0.3829	0.2736	0.2190	0.0719	0.3596	0.2088	0.2845
A2	Sheep manure	0.1556	0.1998	0.1708	0.2113	0.1245	0.1321	0.1967	0.2347
A3	Chicken manure	0.0315	0.1460	0.1010	0.2113	0.1245	0.0759	0.1967	0.1577
A4	Horse manure	0.0296	0.1061	0.1960	0.2113	0.1245	0.1235	0.1967	0.1343
A5	Straw	0.2071	0.0231	0.0242	0.0258	0.4703	0.0299	0.0767	0.0275
A6	Fruit and vegetable waste	0.1372	0.0518	0.0369	0.0524	0.0421	0.0334	0.0726	0.0911
A7	Household food waste	0.0288	0.0904	0.1975	0.0687	0.0421	0.2457	0.0517	0.0702

.

The prioritization with respect to the sub-criteria shows that cattle manure ranks highest in all sub-criteria except storability, where straw is ranked highest due to its ability to be stored for prolonged periods without deterioration or nutrient loss. This partially explains why cattle manure is currently the preferred choice of feedstock for small-scale biogas systems in many developing countries.

The overall weights and ranking of the feedstock alternatives are obtained by summing the weighted products of the corresponding sub-criteria weights and alternatives, generating the weights for each alternative with respect to each criterion, as shown in

Table 11. Overall decision matrix and feedstock ranking.

		Criteria
		C1	C2	C3
		30.92%	58.13%	10.96%	Overall Weight	Ranking
A1	Cattle manure	0.1322	0.0467	0.1093	24.00%	1
A2	Sheep manure	0.0592	0.0486	0.0543	15.75%	3
A3	Chicken manure	0.0296	0.0425	0.0355	11.33%	5
A4	Horse manure	0.0226	0.0508	0.0442	12.41%	4
A5	Straw	0.0384	0.1023	0.0112	21.76%	2
A6	Fruit and vegetable waste	0.0315	0.0140	0.0170	5.91%	7
A7	Household food waste	0.0199	0.0285	0.0620	8.84%	6

. For example, the overall weight of alternative A1 (cattle manure) across the different criteria is calculated as follows: under criterion C1 (adequacy), the prioritization of A1 is determined by summing the product of the relative weight of SC1 (16.67%) and A1’s prioritization under SC1 (0.4102) with the product of the relative weight of SC2 (16.67%) and A1’s prioritization under SC2 (0.3829), resulting in a value of 0.1322. The final overall weighting and ranking of the feedstock alternatives is then calculated using Equation (6).

The final prioritization of the different potential feedstocks is illustrated in Figure 5 below.

From the resulting prioritization in Table 11 and Figure 5, cattle manure is the top-ranked feedstock alternative in the Moroccan region of Fès-Meknès. This is because it is readily digestible in biogas plants, is produced in sufficient quantities, as seen in farms having up to 11 cattle on average, and ease of collection, as most farmers keep their cattle in an enclosure for part or all of the day as reported in a study conducted in the area [30]. Additionally, cattle manure is the most commonly and widely utilised feedstock in small-scale biogas plants; it also contains the right anaerobic bacteria required to produce biogas. Cattle manure ranks higher under both adequacy and logistical ease criteria. Except for extreme occurrences such as severe drought and disease outbreak, manure production from cattle is fairly consistent. Since manure collection is part of the daily routine of maintaining cowshed cleanliness, additional labour is typically not required to supply this manure to a farm-based biogas plant. Proper management of manure in Morocco is a positive step, as it is usually left to decompose for up to a year before being used as organic manure or for soil amendment. This practice, however, risks contaminating water bodies with runoff and leachate from manure piles and releasing methane, a significant greenhouse gas, during decomposition. Using cattle manure as a feedstock for biogas production offers significant benefits, including energy generation from biogas and producing digestate, a more concentrated manure or organic fertilizer.

Straw is ranked as the second most important feedstock. Cereals, particularly wheat and barley, are the most prevalent crops grown in Morocco and the Fès-Meknès region as seen in [30,35]. Wheat, for instance, represents a significant component of the Moroccan population’s nutritional needs, and substantial amounts are harvested in the Fès-Meknès region. Although straw is seasonal and requires more effort to collect, transport, and store, it can be preserved for longer periods without degradation or nutrient loss under the right conditions, which enhances its supply reliability.

Supply consistency, which has the highest weight among all the criteria, is where straw excels among all potential feedstocks. However, its use as livestock feed and for mulching impacts the amount of straw available as biogas plant feedstock. The by-product digestate can be utilized as fertilizer to boost cereal crop production, reducing expenditure on organic fertilizers and improving farmers’ economic conditions. Despite this, straw is not readily digestible in its collected form and may require pretreatment, such as size reduction, before being fed to a biogas plant.

Sheep manure was ranked third because the amount collected per animal is almost ten times lower than that from cows, and sheep/goats typically graze outside for most of the day, making collection difficult. Large herds are required to supply adequate feedstock to a biogas plant. Additionally, their manure is much drier (higher total solids) compared to cattle manure, necessitating more water and perhaps some pretreatment before feeding it into the biogas plant. Horse manure and chicken manure are ranked fourth and fifth, respectively. The number and prevalence of horses is low in the Fès-Meknès region. Chickens produce low amounts of manure per head and with an average number of about 20 chicken per farm, the quantities that can be collected are significantly low. Similar to sheep or goat manure, horse manure is drier than cattle manure and requires more water for use in a biogas plant.

Household food waste is ranked second last. Due to its continuous supply as a by-product of daily nourishment, it can be a reliable addition to the main feedstocks of a biogas plant, enhancing the nutritional content of the feedstock. However, their low quantities per household per day and typical use as animal feed makes it a lowly ranked feedstock.

Fruit and vegetable waste is ranked last. Although these feedstocks have good nutritional content and digestibility, they are seasonal and highly perishable. Additionally, only damaged and or unsold fruits and vegetables can be used as animal feed or fed to a biodigester. However, they can supplement the main feedstocks in a biogas plant whenever available. Based on the discussion and ranking above, cattle manure emerges as a suitable base feedstock for small-scale biogas plants in the Fès-Meknès region. It can be co-digested with straw to enhance biogas production, and sheep manure can also be added as a complementary feedstock.

Co-digestion, commonly implemented in large-scale biogas plants, enhances feedstock nutrient profiles, such as the carbon-to-nitrogen (C/N) ratio, while diversifying feedstock sources to mitigate shortages and boost biogas yields. For example, straw has a very high C/N ratio, exceeding 80, whereas livestock manure from cattle, sheep, chickens, and pigs is rich in nitrogen [44,45]. Combining these feedstocks allows for balancing the C/N ratio to recommended optimal levels of 20–30. The complimentary characteristics of cattle manure and straw alongside the ranking obtained above make cattle manure and straw the ideal feedstock choice for sustainable and resilient small-scale biogas plants in the region of Fès-Meknès.

4. Conclusions and Recommendations

This study applied the Analytic Hierarchy Process (AHP) to biodigester feedstock selection in Morocco, developing a systematic approach to aid decision-making in small-scale biogas adoption. The findings highlight cattle manure, straw, and sheep manure as the most suitable feedstocks, making farms with these resources prime candidates for biogas plant rollout.

While the rankings provide a regional overview, they can be refined for specific farms or farm clusters with minimal deviation. The methodology presented in this study is adaptable and can be replicated in other regions, with the case study demonstrating its applicability in a typical setting. Its flexibility makes it well suited for varying and complex scenarios. When combined with policy support, cost considerations, biogas plant models, and environmental factors, this approach can enhance biogas adoption, improve its reliability as a renewable energy source, and contribute to sustainable development.

Future research should focus on refining the selection criteria (and sub-criteria), optimizing feedstock mixing ratios, and exploring necessary pretreatment methods. Additionally, a techno-economic analysis of co-digestion in small-scale biogas systems could provide deeper insights into its feasibility. These analyses can then inform tailored recommendations for both individual and cooperative-operated biogas plants. Furthermore, policymakers and stakeholders should integrate feedstock selection into biogas promotion strategies, including incentives and support measures to ensure that adopters have access to diverse and sustainable feedstocks for their plants.