Behavioral decision-making and marketing institution preferences in coconut agroindustry development: evidence from SWOT-AHP and value-added analysis in Indonesia

1 Introduction

Coconut-based agroindustry plays an important role in supporting rural livelihoods, employment generation, and local economic resilience in many developing countries, particularly in Southeast Asia. In Indonesia, coconut commodities contribute significantly to household income generation in rural areas through farming, processing, and marketing activities, including value-added downstream products and diversified coconut-based agroindustry (Ghosh, 2015; Mardesci et al., 2021). Empirical evidence from Central Sulawesi demonstrates that coconut farmers retain approximately 70% of the final consumer price through existing marketing channels, although overall marketing efficiency remains relatively modest, ranging from 21% to 26%, indicating substantial inefficiencies within the value chain (Rauf et al., 2023). Similar findings across Indonesian agroindustrial sectors show that direct marketing systems and stronger institutional linkages can improve farmer income shares, reduce dependency on intermediaries, and strengthen local economic resilience (Folinas and Fotiadis, 2017; Madhani, 2016; Said et al., 2021; Sumekar et al., 2018). Nevertheless, many coconut-producing regions in Indonesia, including Buton Regency, continue to rely heavily on traditional marketing systems characterized by household-scale copra production, limited processing diversification, and intermediary-dominated market structures. Under these conditions, producers frequently receive only a small proportion of the final market value due to weak bargaining power, limited market information, and inefficient institutional arrangements (Chatra and Rosi, 2023; Lestari and Premono, 2019).

The persistence of low-value production systems among coconut farmers and processors cannot be fully explained through conventional neoclassical economics, which assumes that economic actors always make perfectly rational and utility-maximizing decisions. In reality, rural agroindustry actors operate under conditions of bounded rationality, incomplete information, institutional uncertainty, and asymmetric market power (Křečková and Brožová, 2017; Manevska-Tasevska and Hansson, 2011). Marketing decisions are often influenced not only by price considerations but also by social trust, transaction security, payment certainty, accessibility, and long-standing institutional relationships with intermediaries and traders (Ferdous and Nunan, 2025; Ferrol-Schulte et al., 2014; Wąs et al., 2021). Studies on Indonesian agroindustry systems reveal that patron–client relationships, information asymmetries, and limited access to price transparency significantly shape farmer bargaining positions and institutional preferences (Chatra and Rosi, 2023; Lestari and Premono, 2019). In many cases, producers continue to prefer familiar marketing institutions despite potentially higher profits available through alternative channels because trusted institutional relationships reduce perceived transaction risk and uncertainty. These patterns align closely with behavioral economics perspectives emphasizing bounded rationality and decision-making under uncertainty rather than purely profit-maximizing behavior (Githinji et al., 2023; Malawska and Topping, 2016; Wuepper et al., 2023).

Previous studies on coconut agroindustry development have primarily focused on marketing efficiency, farmer's share, value-added analysis, and agribusiness profitability using conventional economic approaches, including product diversification, cost calculation, and downstream coconut-processing feasibility (Ghosh, 2015; Mardesci et al., 2021). Research conducted in Indonesian coconut systems has shown that processing activities such as copra production generate important economic value and contribute to rural income diversification (Rauf et al., 2023). Other studies within Indonesian agroindustry contexts demonstrate that direct marketing channels generally provide higher farmer's shares compared to indirect marketing systems involving multiple intermediaries (Sumekar et al., 2018). Additionally, governance structures, certification systems, and institutional arrangements have been identified as important determinants of marketing performance and producer welfare across agricultural commodities (Hutagaung et al., 2024). However, most existing studies remain largely descriptive and production-oriented, focusing primarily on technical efficiency and marketing margins while giving limited attention to the behavioral dimensions underlying institutional preferences and strategic decision-making among agroindustry actors. Behavioral economics research itself has also concentrated predominantly on laboratory experiments and urban consumer behavior, leaving rural agroindustry systems substantially underrepresented within field-based behavioral studies.

This limitation is particularly important because rural agroindustry decision-making occurs within complex institutional environments shaped by trust networks, social obligations, transaction risks, and information constraints that cannot be adequately replicated in laboratory settings (Brock and Barham, 2013; Cordaro and Desdoigts, 2021; Mastenbroek, 2025). Evidence from Indonesian agricultural marketing systems demonstrates that information asymmetry and institutional dependency strongly influence farmer choices regarding marketing channels and strategic production decisions (Chatra and Rosi, 2023; Lestari and Premono, 2019). Furthermore, governance reforms and certification systems such as ISPO-related institutional arrangements in Indonesian agribusiness indicate that institutional structures can substantially alter marketing behavior beyond conventional price incentives (Hutagaung et al., 2024). These findings suggest that agroindustry actors evaluate marketing institutions multidimensionally by considering demand stability, payment certainty, accessibility, transaction security, and institutional trust rather than focusing exclusively on price optimization (Liu et al., 2025; Wąs et al., 2021). Consequently, understanding coconut agroindustry development requires a behavioral economics framework capable of capturing naturally occurring decision-making behavior within real market environments.

This study addresses the following research question: how do economic outcomes, institutional preferences, and behavioral decision mechanisms influence coconut-agroindustry development strategies in rural market environments? Specifically, the study examines how actors prioritize the criteria used to evaluate marketing institutions and how economic and institutional evidence shapes strategic development options under market uncertainty, information limitations, and institutional constraints. Drawing on behavioral economics and bounded-rationality perspectives, the study proposes that producers evaluate marketing arrangements through multidimensional considerations involving price, payment certainty, accessibility, demand stability, product absorption, and transaction security rather than through profit maximization alone.

To examine these behavioral dynamics, this study employs a field-based behavioral approach because field analysis provides stronger external validity than laboratory experiments in explaining real-world economic behavior. Laboratory experiments often isolate participants from the actual institutional dependencies, financial risks, and social relationships that shape decision-making behavior in rural agroindustry systems. In contrast, field-based analysis allows economic actors to express preferences and strategic choices within naturally occurring market ecosystems where institutional trust, transaction costs, and market uncertainty operate simultaneously. Previous field-based studies in Indonesian agricultural marketing systems demonstrate that real market environments reveal behavioral patterns and institutional constraints that are often absent in controlled laboratory conditions (Chatra and Rosi, 2023; Malawska and Topping, 2016; Permana et al., 2020; Rauf et al., 2023; Wuepper et al., 2023). Therefore, field analysis is particularly appropriate for examining coconut agroindustry behavior because marketing decisions are deeply embedded within local social structures and institutional arrangements.

The study was conducted in major coconut-producing districts of Buton Regency and involved 489 copra processors, 182 coconut shell charcoal processors, marketing intermediaries, and institutional stakeholders operating under naturally occurring market conditions. The empirical analysis integrated value-added calculation, farmer's share, marketing margins, a price-spread efficiency index, ordinal preference weighting, and group-level SWOT-AHP strategy synthesis. This sequence connected measured economic outcomes with the institutional criteria and strategic groups identified in the field.

This study contributes to the behavioral economics and agroindustry literature in three ways. First, it extends preference-based behavioral analysis to a rural agroindustry setting in which decisions are embedded in institutional dependence, liquidity constraints, and incomplete market information. Second, it integrates observed economic outcomes—value added, marketing margins, farmer's share, and channel efficiency—with transparent ordinal criterion weighting and group-level SWOT-AHP strategic prioritization. Third, it links the criteria used in the decision framework to behavioral interpretation: price represents monetary payoff, demand stability reflects uncertainty reduction, accessibility captures transaction and cognitive frictions, payment speed represents liquidity certainty, and product absorption capacity reflects transaction-completion security. Because these constructs were not measured with validated psychometric scales, the behavioral contribution is interpreted as preference-based and conceptual rather than as a latent-variable test.

The remainder of this article is organized as follows. The next section explains the field-based study design, participant selection procedures, data collection techniques, and analytical methods employed in the study. The following section presents empirical findings regarding value-added generation, marketing efficiency, institutional preferences, and strategic development priorities in coconut agroindustry systems. The final section discusses the behavioral economics implications of the findings, policy recommendations, study limitations, and future research directions.

2 Literature review

2.1 Behavioral economics and rural decision-making

Behavioral economics has increasingly been recognized as an important approach for understanding economic decision-making in rural agroindustry systems, particularly in contexts characterized by uncertainty, limited information, and institutional dependency. Unlike neoclassical economics, which assumes that individuals always behave rationally to maximize utility, behavioral economics emphasizes that economic actors often operate under bounded rationality, where decisions are constrained by limited cognitive capacity, incomplete information, and contextual pressures. In rural agroindustry environments, producers frequently prioritize familiar institutional arrangements, transaction certainty, and trust-based relationships rather than purely profit-maximizing outcomes. Studies have shown that rural producers tend to simplify complex market decisions by relying on heuristics and socially embedded transaction patterns, especially when faced with volatile market conditions and weak institutional support systems (Costanigro et al., 2011; Saaty, 2008). These behavioral tendencies demonstrate that economic decisions in agroindustry systems are strongly influenced by psychological and institutional considerations (Githinji et al., 2023; Křečková and Brožová, 2017; Manevska-Tasevska and Hansson, 2011; Wuepper et al., 2023).

Institutional trust and transaction security also play a crucial role in shaping marketing behavior within rural agroindustry systems. Marketing relationships in agricultural commodity chains are often embedded within long-standing social interactions between producers, traders, and intermediaries, including patron-client arrangements that shape access to finance, markets, and transaction security (Ferdous and Nunan, 2025; Ferrol-Schulte et al., 2014; Noak, 2024). These relationships provide non-price benefits such as flexible transactions, informal credit access, immediate cash payments, and reduced market uncertainty. Consequently, producers may continue to rely on intermediary-based marketing systems despite lower economic returns because these institutions offer greater perceived transaction security and stability. Research on rural commodity markets in Indonesia indicates that patron–client relationships and institutional asymmetry strongly influence bargaining power and market participation among producers (Lestari and Premono, 2019). Furthermore, transparency mechanisms and trust-enhancing institutional arrangements have been shown to improve market participation and reduce opportunistic behavior within agricultural marketing systems (Modekurti, 2016). These findings highlight the importance of behavioral and institutional dimensions in understanding rural market dynamics, especially where trust deficits raise transaction costs and weaken participation in formal market systems (Mastenbroek, 2025; Wąs et al., 2021).

Behavioral economics is therefore highly relevant for analyzing strategic decision-making in rural agroindustry development because it captures the interaction between cognitive behavior, institutional structures, and market uncertainty. Strategic development decisions in agroindustry systems are not determined solely by economic profitability but are also influenced by perceptions of risk, institutional familiarity, and social trust. The use of multicriteria decision-making approaches such as Analytical Hierarchy Process (AHP), SWOT-AHP, and other MCDA frameworks has become increasingly important in behavioral agroindustry studies because these methods allow intangible behavioral factors to be incorporated into strategic evaluation processes (Ayswer et al., 2024; Basar, 2024; Bruzzese et al., 2023; Grošelj et al., 2024; Nugraha et al., 2022; Phadermrod et al., 2019). Through these approaches, agroindustry planning can better reflect real-world producer behavior and generate more realistic development strategies compared with purely technical or economic analyses.

2.2 Coconut agroindustry and marketing efficiency

Coconut agroindustry development contributes significantly to rural economic growth through value-added processing activities capable of increasing household income, employment opportunities, and agroindustrial diversification. The transformation of coconut products into copra, charcoal, coconut oil, coconut sugar, coconut water, beverages, and other derivative products enables producers to capture greater economic value within the supply chain (Cunha Júnior et al., 2023; Mardesci et al., 2021). Previous studies on coconut agroindustry systems have demonstrated that downstream processing activities strengthen local economic resilience by expanding value chains and increasing marketable product diversification. Research conducted in Central Sulawesi found that coconut processing systems generated substantial marketing margins and farmer's shares approaching 70%, indicating that value-added processing activities contribute positively to local economic performance (Rauf et al., 2023). Similar findings in other agroindustry sectors also emphasize that processing-based value chains create stronger economic multipliers compared with raw commodity marketing systems (Agus et al., 2025; Chatra and Rosi, 2023; Elfahmi et al., 2024; Ismail et al., 2026).

Marketing efficiency within coconut agroindustry systems is strongly influenced by marketing channel structure, institutional arrangements, information accessibility, and the integration of marketing functions with supply-chain coordination (Bheekharry, 2021; Flint, 2004; Folinas and Fotiadis, 2017; Išoraitė, 2024). Marketing channels involving multiple intermediaries often generate higher cumulative marketing margins while reducing farmer's share and weakening producer bargaining power. Conversely, shorter marketing channels tend to improve efficiency because fewer transaction costs accumulate throughout the distribution process. Previous studies indicate that marketing efficiency is closely associated with channel length, transportation costs, access to market information, and institutional coordination among market actors (Madhani, 2016; Paramita et al., 2020; Rauf et al., 2023; Said et al., 2021). Moreover, institutional structures such as farmer cooperatives and organized producer groups have been shown to improve market access and reduce marketing asymmetry within agricultural commodity systems. These findings suggest that efficient marketing systems require both structural improvements and institutional strengthening to support more equitable value distribution among agroindustry actors.

Despite the economic advantages associated with shorter marketing channels and direct market access, many coconut producers continue to rely heavily on intermediaries and traditional trading arrangements. This persistence reflects the existence of institutional dependency and behavioral lock-in effects within rural commodity markets. Intermediaries often provide critical non-price services such as immediate cash payments, transportation assistance, market access facilitation, and informal financial support, which remain highly valuable for small-scale producers facing liquidity constraints and market uncertainty (Ferdous and Nunan, 2025; Ferrol-Schulte et al., 2014; Noak, 2024). Studies on agricultural marketing systems have shown that weak price information, unequal bargaining power, and entrenched trader–producer relationships contribute to the continued dominance of intermediary-based marketing systems (Chatra and Rosi, 2023; Lestari and Premono, 2019). Consequently, improving marketing efficiency in coconut agroindustry systems requires not only structural market reforms but also institutional interventions capable of replacing the transaction security traditionally provided by intermediaries.

2.3 Strategic development and institutional strengthening

Strategic development planning in agroindustry systems increasingly utilizes integrated multicriteria decision-making approaches such as SWOT-AHP to improve the objectivity and consistency of strategic evaluation. SWOT analysis enables researchers and policymakers to identify internal strengths and weaknesses alongside external opportunities and threats, while Analytical Hierarchy Process (AHP) allows these strategic factors to be quantified through pairwise comparison procedures. The integration of SWOT and AHP within the SWOT-AHP framework therefore provides a systematic approach for prioritizing strategic alternatives based on measurable relative importance (Basar, 2024; Bruzzese et al., 2023; Dimitrova and Desev, 2018; Nikolić et al., 2015; Saaty, 2008; Solangi et al., 2019). Previous studies have shown that SWOT-AHP is highly effective for agroindustry planning because it transforms qualitative stakeholder perceptions into quantitative strategic priorities that support evidence-based policy formulation (Ayswer et al., 2024; Grošelj et al., 2024; Mukul and Güler, 2024; Nugraha et al., 2022; Phadermrod et al., 2019; Stoilova and Martinov, 2019).

Several internal and external factors have been identified as dominant determinants of agroindustry development strategies. Internal factors commonly include resource availability, production capacity, institutional capability, infrastructure quality, and market accessibility, whereas external factors involve market demand, policy support, competition, and environmental uncertainty. In coconut agroindustry systems, strategic planning is particularly influenced by marketing structure, farmer bargaining power, value-added diversification opportunities, and institutional coordination mechanisms. Previous research has consistently emphasized that access to market information and institutional support strongly shape strategic development outcomes in rural agroindustry sectors (Rauf et al., 2023). These findings indicate that sustainable agroindustry development requires integrated strategic planning capable of addressing both internal production constraints and external market dynamics simultaneously.

Institutional strengthening is also considered essential for improving competitiveness and sustainability within rural agroindustry systems. Strong institutions enhance transaction transparency, improve bargaining power, facilitate market access, and reduce uncertainty among agroindustry actors by replacing informal dependency with more accountable coordination mechanisms (Ferdous and Nunan, 2025; Ferrol-Schulte et al., 2014; Noak, 2024). Institutional support mechanisms such as cooperatives, marketing associations, and extension services can help producers access better market information, strengthen collective bargaining positions, and improve participation within formal market systems. Studies have demonstrated that institutional transparency and trust-based governance structures significantly improve market performance and reduce opportunistic behavior in agricultural commodity systems (Lestari and Premono, 2019; Modekurti, 2016). Therefore, strengthening rural institutions is not only an economic necessity but also a behavioral intervention capable of encouraging innovation adoption, improving strategic confidence, and supporting long-term agroindustry sustainability.

2.4 Alternative criteria-weighting methods and hybrid SWOT-MCDM approaches

Criteria-weighting methods in MCDM can be broadly classified as subjective, objective, and hybrid approaches. Subjective methods derive criterion importance from expert judgments and cognitive evaluations, whereas objective methods infer weights from the statistical, dispersion, or information-theoretic properties of an observed decision matrix. Hybrid approaches combine both sources to balance decision-maker preferences with data-driven structure. Recent systematic reviews emphasize that method selection should be aligned with the decision context, the availability and quality of performance data, the number of criteria, the required level of stakeholder involvement, and the degree of uncertainty in the judgments or measurements (Sarkar et al., 2026a; Sarkar and Goswami, 2026). This classification is important in the present study because the marketing-preference and SWOT components are fundamentally expert-informed rather than alternative-performance matrices from which purely objective weights could be estimated.

Within the subjective class, AHP remains a justified choice when the hierarchy is relatively small, the decision problem requires explicit expert prioritization, and transparency of the judgment process is important. Its principal advantages are the intuitive hierarchical decomposition of a complex problem, reciprocal pairwise comparisons, directly interpretable ratio-scale priorities, and an explicit consistency diagnostic that allows researchers to evaluate the logical coherence of judgments. AHP is also widely understood by practitioners and can facilitate structured discussion among heterogeneous stakeholders. Its limitations include the n(n−1)/2 comparison burden as the number of criteria increases, sensitivity to subjective scale use, the possibility of inconsistent judgments, and potential rank-reversal concerns. Reduced-comparison methods address several of these limitations: FUCOM requires only n−1 comparisons and seeks full consistency; LBWA organizes criteria into importance levels with a lower elicitation burden; and DIBR emphasizes relationships among ranked, particularly adjacent, criteria. These newer approaches can be advantageous for larger or cognitively demanding hierarchies, although they still depend on expert ordering or preference information (Pamučar et al., 2018; Sarkar and Goswami, 2026; Tešić et al., 2022; Žižović and Pamučar, 2019).

AHP was retained for the four-group SWOT component because the hierarchy contained only strengths, opportunities, weaknesses, and threats, and the retained study record already provided a normalized group-priority vector. In this setting, the small number of groups limits the pairwise-comparison burden, while the ratio interpretation and consistency framework remain useful for transparent strategic synthesis. Objective weighting methods were not adopted for these SWOT groups because their importance represents expert strategic judgments rather than dispersion in an alternative-by-criterion performance matrix. Thus, AHP is methodologically defensible here not because it is universally preferable to newer methods, but because its assumptions match the structure and information source of this specific decision problem.

The marketing-institution criterion weights require a different interpretation. The original study procedure used AHP, but the archived record retained only the final ordinal ordering of the five criteria and not the underlying pairwise-comparison matrices. The revised manuscript therefore uses the Rank-Order Centroid (ROC) method to avoid reconstructing unsupported expert judgments. This methodological change preserves the observed ranking but imposes a rank-derived spacing among criteria; consequently, ROC cannot recover the original intensity of pairwise preferences, expert-level disagreement, or AHP consistency ratios. The cardinal magnitudes of the reported criterion weights could therefore differ from weights that would have been obtained from the original AHP matrices, especially if the original judgments contained unusually strong or weak preference intensities. The present weights should accordingly be interpreted as transparent, order-preserving estimates for criterion prioritization rather than as equivalent reconstructions of the original AHP eigenvector. The absence of rank reversals in the subsequent perturbation tests supports the stability of the retained ordering, but it does not restore the information lost with the unavailable pairwise matrices.

SWOT has also been combined with a wide range of MCDM techniques. Recent applications include SWOT–Fuzzy SIWEC for prioritizing artificial-intelligence contributions to poverty eradication, integrated SWOT-AHP for shared-mobility services, and broader strategic-management applications emphasizing the role of SWOT in long-term organizational sustainability (Badi and Bayane Bouraima, 2026; Dobrodolac et al., 2024; Farag, 2025). Hybrid studies have further combined SWOT with TOPSIS, fuzzy approaches, and other intelligent decision-support models. Contemporary reviews likewise show a movement toward hybrid weighting, uncertainty-aware modeling, cross-method comparison, and explicit robustness assessment rather than reliance on a single deterministic weighting or ranking procedure (Nedeljkovic et al., 2026; Sarkar et al., 2026a; Sarkar and Goswami, 2026). Positioned within the subjective-objective-hybrid taxonomy introduced above, SWOT-Fuzzy SIWEC and fuzzy SWOT-TOPSIS applications are hybrid in the sense that they combine subjective expert judgment on SWOT factors with an objective or uncertainty-modeling layer for aggregation, whereas the SWOT-AHP applications cited here remain subjective throughout, since both the SWOT-factor identification and the weighting step rely on expert judgment rather than measured alternative-performance data. The present study falls into this latter, purely subjective category, which is the reason a data-driven objective weighting method could not substitute for the missing AHP pairwise matrices.

2.5 Conceptual framework and scope of behavioral interpretation

The conceptual framework presented in Figure 1 integrates two empirical evidence streams. First, value-added and marketing-channel analyses quantify the economic incentives and structural constraints faced by coconut processors. Second, the preference criteria and SWOT groups organize the institutional considerations that shape strategic choices. Selling price represents monetary payoff; demand stability reflects uncertainty reduction; accessibility captures transaction frictions; payment speed represents liquidity certainty; and product absorption capacity reflects transaction-completion security. These indicators are interpreted as observable preference mechanisms rather than as psychometrically validated latent variables.

3 Materials and methods

3.1 Context and field setting

This study employed a field-based behavioral economics approach conducted within the naturally occurring environment of the coconut agroindustry system in Buton Regency, Southeast Sulawesi, Indonesia. The research was carried out from September to December 2024 in major coconut-producing districts where copra processing and coconut shell charcoal agro-industries operated under normal market conditions. Buton Regency was purposively selected because it represents an important coconut production center characterized by traditional agroindustry systems, multilayer marketing channels, and strong institutional interactions among farmers, collectors, traders, and processors. Previous studies on Indonesian coconut marketing systems demonstrated that coconut value chains involve multiple institutional actors and marketing structures that significantly influence bargaining power, market participation, and economic efficiency (Lestari and Premono, 2019; Rauf et al., 2023).

The naturally occurring market conditions enabled behavioral responses and institutional preferences to emerge through routine production and marketing activities without artificial market manipulation. Earlier agricultural marketing studies showed that field-based observations provide stronger ecological validity because behavioral responses are generated within authentic economic environments involving real transaction risks, market uncertainty, and institutional dependency (Ahmadi et al., 2021; Sumekar et al., 2018). The field setting therefore provided an appropriate context for examining strategic decision-making, institutional preferences, and marketing behavior among coconut agroindustry actors operating in real-world rural markets.

3.2 Participants and sampling

The participants consisted of actors involved in the coconut value chain, including coconut farmers, copra processors, coconut shell charcoal producers, traders, marketing intermediaries, and institutional stakeholders operating in seven coconut-producing districts in Buton Regency. Participants were selected using purposive sampling to ensure representation of both agroindustry specialization and marketing structure diversity. Respondents were included if they had actively operated coconut-related businesses for at least two consecutive years, possessed complete production and financial records, and were directly involved in production or marketing decision-making activities. Similar purposive approaches have been widely applied in Indonesian agroindustry and marketing studies to capture institutional and behavioral diversity within agricultural value chains (Chatra and Rosi, 2023; Rauf et al., 2023).

The final sample consisted of 489 copra processors, 182 coconut shell charcoal processors, marketing actors, and 10 institutional experts representing government agencies, agricultural extension services, academics, and farmer organizations. Institutional stakeholders were included because previous studies demonstrated that institutional arrangements strongly influence marketing efficiency, bargaining power, and channel governance in rural agricultural systems (Lestari and Premono, 2019). Statistical adequacy was confirmed using G*Power 3.1 with α = 0.05, statistical power = 0.80, and medium effect size (f² = 0.15), indicating that the minimum recommended sample size was 146 respondents and confirming that the final sample size was statistically sufficient.

3.3 Study design and behavioral preference elicitation

The study used a cross-sectional, field-based mixed-method design to examine economic performance, marketing structures, institutional preferences, and strategic development priorities under naturally occurring market conditions. No prices, payment schedules, or market relationships were experimentally manipulated. Structured interviews and field observations generated the economic and institutional evidence, while the expert panel identified and ordered the marketing criteria and assessed the four SWOT groups.

The preference-elicitation stage focused on five criteria: selling price, demand stability, accessibility, payment speed, and product absorption capacity. The retained study record preserved their final ordinal order but not the individual pairwise-comparison matrices. The revised analysis therefore converts the order into transparent rank-order centroid weights and treats the criteria as preference proxies. The strategic component uses the archived normalized SWOT-group vector for strengths, opportunities, weaknesses, and threats.

3.4 Data collection and variables

Data were collected from primary and secondary sources to capture economic outcomes and decision preferences. Primary data were obtained through structured interviews, direct field observations, institutional assessments, and structured criterion-ordering and SWOT-group evaluation exercises involving farmers, traders, processors, and institutional stakeholders. Secondary data were collected from regional agricultural agencies, trade offices, Statistics Indonesia (BPS), and previous agroindustry reports. Combining interviews, observations, marketing records, and documentary evidence supported contextual triangulation.

The principal outcomes were value added, the value-added ratio, profitability, marketing margins, farmer's share, the price-spread efficiency index, normalized criterion weights, SWOT-group weights, and normalized strategy scores. Contextual variables included agroindustry type, district, transport access, market distance, payment arrangements, demand conditions, and institutional relationships. Analytical procedures were implemented in Python and verified through independent recalculation of all reported weights and sensitivity scenarios.

3.5 Ethics and informed consent

This study received ethical approval from the Institutional Research Ethics Committee of the affiliated university under ethical clearance number 021/KEP-AGR/IX/2024. Ethical approval was required because the study involved human participants engaged in field-based behavioral evaluations conducted within naturally occurring market environments. Previous behavioral market studies emphasized the importance of ethical oversight to ensure participant protection, confidentiality, and responsible data collection practices in real-world economic settings (Costanigro et al., 2011).

All participants voluntarily agreed to participate after receiving information about the study objectives, confidentiality, data anonymization, and their right to withdraw. To reduce hypothesis-guessing and social-desirability effects, participants were informed about the general decision-making purpose but not the expected ordering of the marketing criteria or SWOT groups before completing the elicitation tasks. No deceptive payment, price, or market manipulation was implemented. Participants received a full explanation of the analytical purpose after data collection.

3.6 Analytical procedures

The analysis was conducted sequentially. Value added was calculated using the Hayami framework as VA=OV−IC−RMV, where VA is value added per kilogram of raw material, OV is output value, IC is the value of intermediate and supporting inputs, and RMV is raw-material value. The value-added ratio was VAR=(VA/OV)×100. Farmer's share was FS=(Pf/Pc)×100, where Pf is the price received by farmers or processors and Pc is the final selling price. Marketing margin was MM=Pc−Pf. The marketing price-spread efficiency index was MEI=(MM/Pc)×100; lower MEI values indicate a shorter and more efficient distribution structure.

Because the retained marketing-preference output contained an ordinal consensus rather than pairwise matrices, criterion weights were estimated with the rank-order centroid method. For n criteria and criterion rank r, the normalized weight is w(r)=(1/n)×[1/r+1/(r+1)+…+1/n]. With n=5, this generated weights of 0.4567, 0.2567, 0.1567, 0.0900, and 0.0400 for selling price, demand stability, accessibility, payment speed, and product absorption capacity, respectively. This method preserves the reported order without presenting reconstructed values as original AHP eigenvector estimates.

For the SWOT-AHP component, a positive reciprocal matrix A=[a(i,j)]satisfiesa(i,j)>0,a(j,i)=1/a(i,j),anda(i,i)=1. The priority vector w satisfies Aw=λmaxw and is normalized so that the sum of w(i) equals 1. Consistency is assessed by CI=(λmax−n)/(n−1) and CR=CI/RI, with RI=0.90 for n=4 and an acceptance threshold of CR<0.10. The retained analytical record contained the normalized consensus vector w=(0.4658,0.2771,0.1611,0.0960), rather than the ten individual matrices. For reproducibility of the reported vector, the equivalent reciprocal matrix was generated as a*(i,j)=w(i)/w(j). This equivalent matrix reproduces the retained weights exactly and is perfectly consistent by construction (λmax=4.0000, CI=0.0000, CR=0.0000); it is not presented as a substitute for the unavailable individual expert matrices.

The SWOT-AHP synthesis followed three stages. First, strengths, weaknesses, opportunities, and threats were identified from interviews, field observations, value-added results, and marketing-channel evidence. Second, the retained group weights were normalized to sum to one. Third, group-level TOWS strategy scores were calculated as qSO=wS+wO,qST=wS+wT, qWO=wW+wO,andqWT=wW+wT. Because every group enters two combinations, the four raw scores sum to 2; final strategy weights were therefore Sq=q/2. The resulting normalized priorities were used to rank SO, ST, WO, and WT strategies.

The economic evidence entered the decision framework through an explicit evidence map rather than by mechanically converting percentages into subjective weights. High value added and profitability informed strengths and opportunities related to diversification and waste valorization. Limited adoption, technological constraints, and infrastructure gaps informed weaknesses. Price volatility, market uncertainty, and climate exposure informed threats. Marketing-channel results motivated the preference criteria by showing the economic benefit of shorter channels and the continuing institutional value of rapid payment, accessibility, stable demand, and transaction completion.

Validation and sensitivity analysis serve related but distinct purposes in MCDM. Sensitivity analysis evaluates whether rankings or priorities remain stable when inputs or assumptions are perturbed, whereas validation assesses whether the decision model is internally coherent and externally credible. Recent reviews distinguish deterministic weight-variation tests, scenario-based analysis, probabilistic or Monte Carlo simulation, fuzzy or interval sensitivity analysis, statistical rank-agreement tests, expert or benchmark validation, and cross-method triangulation as complementary approaches to robustness assessment (Sarkar et al., 2026b). In the present study, the available archived data support a transparent internal robustness assessment rather than full cross-method validation. Each marketing-criterion weight and each SWOT-group weight was therefore multiplied by 0.80, 0.90, 1.10, and 1.20 in separate one-at-a-time scenarios, after which the complete vector was renormalized to sum to one. Criterion ranks and strategy scores were recalculated for every scenario. Robustness was assessed from minimum and maximum weights, rank reversals, and Spearman rank correlation with the baseline order. This procedure tests local ranking stability under controlled weight changes, while qualitative and documentary triangulation provides contextual plausibility. Because the original individual AHP pairwise matrices are unavailable, consistency re-estimation, cross-method comparison based on the original judgments, and probabilistic validation cannot be performed retrospectively and are treated as directions for future research. Missing survey responses were handled by listwise deletion.

4 Results

4.1 Value added and development potential of the coconut agroindustry

The coconut agroindustry in Buton Regency demonstrates substantial economic potential through the processing of coconut-derived products, particularly copra and coconut shell charcoal. Coconut processing activities remain dominated by household-scale agroindustries operating under traditional production systems where copra functions as the primary commercial commodity, while coconut shell charcoal represents a secondary value-added product derived from processing waste. The integration of these processing activities strengthens rural economic resilience because coconut processing generates additional value within the agricultural supply chain and contributes directly to local income generation through processing and marketing activities (Rauf et al., 2023). Similar patterns in Indonesian agroindustrial systems indicate that processing diversification and value-chain upgrading can strengthen rural livelihoods and increase economic opportunities among small-scale producers (Agus et al., 2025; Elfahmi et al., 2024; Ghosh, 2015; Ismail et al., 2026; Mardesci et al., 2021; Nugraha et al., 2022; Sumekar et al., 2018).

The utilization of coconut shell waste for charcoal production is particularly important because it expands the economic value of previously underutilized by-products and supports agroindustrial sustainability through waste-to-value strategies. Although previous coconut studies have not specifically quantified coconut shell charcoal processing, broader agroindustry literature demonstrates that processing diversification and waste utilization contribute significantly to rural industrial sustainability and economic resilience (D’Almeida and de Albuquerque, 2025; Gokulsidhardh et al., 2026; Nugraha et al., 2022; Phadermrod et al., 2019; Singh et al., 2024; Sun et al., 2020). The integration between copra processing and coconut shell charcoal production therefore creates opportunities for value-added diversification, increased household income, and improved resource efficiency within rural production systems. The detailed value-added calculations are presented in Table 1.

Table 1. Value-added calculation results of coconut shell charcoal processing in buton regency.

The results presented in Table 1 indicate that coconut shell charcoal processing generated substantial economic value-added across all observed districts. The average value-added reached IDR 1,002.52/kg with an average value-added ratio of 54.87%, indicating that more than half of the final product value originated from processing activities rather than from raw material value alone. These findings confirm that coconut shell charcoal processing functions as an economically productive value-added activity capable of strengthening the coconut agroindustry value chain. Similar interpretations of value-added generation have also been applied in broader Indonesian agroindustrial studies where processing margins and farmer's shares are used to evaluate economic upgrading within commodity systems (Mardesci et al., 2021; Setyowati et al., 2021; Sumekar et al., 2018). The relatively high value-added ratio further suggests that coconut shell charcoal agroindustry possesses strong potential for sustainable agroindustrial diversification and rural economic expansion.

The production efficiency indicators further demonstrate that the processing system operated with relatively efficient raw material conversion and labor utilization. The average conversion factor of 0.31 indicates that approximately 0.31 kg of charcoal was produced from each kilogram of coconut shell input. Meanwhile, the labor coefficient remained relatively low at 0.0019 HOK/kg of raw material, suggesting that charcoal processing depended more heavily on operational efficiency and raw material management than on intensive labor utilization. Existing coconut marketing studies similarly indicate that agroindustrial profitability is strongly influenced by processing efficiency, market structure, and operational cost management rather than labor intensity alone (Rauf et al., 2023). The dominance of profits within the value-added distribution structure was reflected in the average profit level of IDR 983.13/kg and the average profit ratio of 98.06%, indicating that most generated value-added accumulated as entrepreneurial profit rather than labor compensation.

Despite its strong economic performance, coconut shell charcoal processing remained geographically concentrated within only several districts in Buton Regency. This condition indicates that barriers to agroindustry expansion are not solely associated with raw material availability but are also influenced by technological limitations, institutional constraints, and behavioral hesitation toward adopting more complex processing systems, including biotechnology, digital prediction, and advanced processing technologies that remain unevenly diffused in rural agroindustry contexts (Deng et al., 2025; Mungklachaiya and Salaiwarakul, 2025; Zhang et al., 2024). Studies on Indonesian agroindustry systems show that limited automation, weak institutional support, and information asymmetry often constrain agroindustrial innovation and technology diffusion among rural producers (Lestari and Premono, 2019; Modekurti, 2016; Nugraha et al., 2022). Many coconut processors therefore continue to prioritize copra production because it represents a more familiar, institutionally secure, and operationally simpler economic activity.

From a behavioral economics perspective, this finding reflects the influence of bounded rationality and perceived operational uncertainty on agroindustry diversification decisions. Small-scale producers frequently avoid adopting new processing systems despite potentially higher profitability because innovation adoption involves uncertainty, information limitations, and increased operational complexity (Brock and Barham, 2013; Cordaro and Desdoigts, 2021; Mastenbroek, 2025). Behavioral studies further suggest that cognitive limitations, risk perception, and institutional dependency significantly shape strategic decision-making among rural economic actors (Costanigro et al., 2011; Lestari and Premono, 2019). Consequently, future agroindustry development strategies should prioritize technology dissemination, institutional strengthening, market transparency, and producer assistance programs to encourage wider adoption of coconut shell charcoal processing within rural coconut production systems.

4.2 Marketing system efficiency and institutional structure of the coconut agroindustry

This section discusses the marketing structure of copra and coconut shell charcoal in Buton Regency, focusing on marketing channels, marketing efficiency, farmer's share, institutional preferences, and strategic marketing behavior among agroindustry actors. The analysis demonstrates how marketing structures and institutional arrangements influence economic outcomes, bargaining power, and behavioral decision-making within rural coconut agroindustry systems.

4.2.1 Copra marketing channels

The observed copra marketing pathways are presented in Figure 2.

Figure 2 illustrates the structure of copra marketing channels operating in Buton Regency. Two dominant marketing channels were identified. Marketing Channel I involved farmers, collectors, wholesalers, and industrial buyers, while Marketing Channel II enabled farmers or processors to sell directly to wholesalers before products reached industrial buyers. Similar two-channel marketing structures have also been identified within coconut marketing systems in Central Sulawesi, where collector-based and direct wholesaler channels significantly influenced marketing margins, farmer's shares, and value distribution within the supply chain (Folinas and Fotiadis, 2017; Madhani, 2016; Rauf et al., 2023; Said et al., 2021).

The comparison between both channels demonstrates that shorter marketing structures generated higher marketing efficiency and improved producer bargaining positions. Marketing Channel II reduced cumulative intermediary costs and increased the proportion of the final price received by farmers and processors. Previous studies similarly indicate that shorter marketing chains and direct marketing arrangements generally improve farmer income shares and strengthen value capture within Indonesian agroindustry systems (Sumekar et al., 2018). The reduction of intermediary involvement therefore contributes directly to more equitable economic distribution and greater marketing efficiency within rural coconut markets.

Nevertheless, despite the economic advantages of shorter marketing channels, many producers continued to depend on collectors because collectors provided immediate cash payments, informal credit access, supply aggregation, and more flexible transaction arrangements. The continued reliance on intermediary actors reflects the importance of institutional trust, transaction security, and social relationships within rural marketing systems, including patron-client ties that often compensate for weak formal market institutions (Ferdous and Nunan, 2025; Ferrol-Schulte et al., 2014; Noak, 2024). Studies on Indonesian agricultural value chains demonstrate that patron–client relationships and institutional trust strongly influence marketing behavior and institutional preferences among small-scale producers (Lestari and Premono, 2019). Consequently, strengthening direct market access alone may not sufficiently alter producer behavior unless accompanied by institutional mechanisms capable of replacing the non-price benefits traditionally provided by intermediaries.

4.2.2 Marketing efficiency and marketing margin

The analysis of marketing margins and farmer's share demonstrates important differences in efficiency between intermediary-based and direct marketing channels. Marketing channels involving multiple intermediaries generated higher cumulative marketing margins and lower farmer's shares, indicating less efficient value distribution within the supply chain. In contrast, direct marketing channels reduced intermediary costs and increased the proportion of consumer prices received by farmers and processors. Similar findings within Indonesian coconut marketing systems indicate that channel structure strongly influences marketing efficiency, value distribution, and producer welfare (Bheekharry, 2021; Išoraitė, 2024; Paramita et al., 2020; Rauf et al., 2023; Sabur et al., 2021; Said et al., 2021). The channel-level numerical results are presented in Table 2.

Table 2. Farmer's share and copra marketing efficiency in Buton regency.

Table 2 shows that Marketing Channel I generated a marketing margin of IDR 3,500/kg, a farmer's share of 75.00%, and a price-spread index of 25.00%. Marketing Channel II reduced the margin to IDR 2,000/kg, increased farmer's share to 85.71%, and lowered the price-spread index to 14.29%. Because lower price spread indicates less value leakage between producer and final buyer, Channel II was the more efficient distribution structure.

The relationship between marketing costs and intermediary profits further indicates that intermediaries possessed relatively strong bargaining positions due to their control over transportation systems, market information, buyer access, and transaction coordination. Studies on Indonesian agricultural marketing systems demonstrate that information asymmetry and institutional dependency strengthen intermediary market power while limiting producer bargaining capacity (Chatra and Rosi, 2023; Ferdous and Nunan, 2025; Ferrol-Schulte et al., 2014; Lestari and Premono, 2019; Noak, 2024). Consequently, high marketing margins reflect not only operational costs but also institutional market control within the coconut trading system.

The findings further suggest that improving marketing efficiency requires institutional strengthening among producers through cooperative marketing systems, collective bargaining institutions, and improved access to market information. Institutional interventions emphasizing transparency, price information systems, and marketing coordination have been shown to reduce transaction uncertainty and improve producer bargaining power within Indonesian agroindustry systems (Lestari and Premono, 2019; Modekurti, 2016). Therefore, strengthening producer organizations and market-access institutions represents an essential strategy for improving marketing efficiency and reducing excessive intermediary dependency within the coconut agroindustry.

4.2.3 Marketing-Institution preference criteria

The final criterion-priority structure is presented in Figure 3 and Table 3.

Table 3. Marketing-institution preference criteria and sensitivity results.

Figure 3 presents the five preference criteria in their final order. The revised model intentionally does not rank Sales Agent Lasalimu, Warehouse 88 City Baubau, and Collector Trader because the source record did not retain the alternative-level matrices required for a reproducible comparison. The analysis therefore concentrates on the relative importance of the criteria that producers and experts use when evaluating marketing institutions.

The rank-order centroid analysis assigned the highest weight to selling price (0.4567), followed by demand stability (0.2567), accessibility (0.1567), payment speed (0.0900), and product absorption capacity (0.0400). Selling price and demand stability jointly accounted for 71.33% of the total weight, showing that immediate economic returns and predictable transactions dominated the preference structure. Accessibility remained substantively important because it lowers transport, search, and coordination costs. This concentration should be read cautiously: because rank-order centroid weights decline as a fixed harmonic function of rank position rather than as an estimate of elicited preference intensity, part of the apparent dominance of the top two criteria is a property of the ROC formula itself and not solely evidence of how strongly experts differentiated between criteria. The ordinal dominance of selling price and demand stability is well supported by the retained consensus ranking, but the specific magnitude of 71.33% should not be interpreted as equivalent to an AHP-derived intensity measure.

Payment speed reduced liquidity exposure, while product absorption capacity represented the likelihood that output could be sold without delay or rejection. These criteria connect the economic and behavioral components of the study, but they should be interpreted as observable preference proxies rather than validated measurements of trust, risk aversion, or bounded rationality.

The sensitivity analysis confirmed the stability of the ordinal preference structure. Across all 20 one-at-a-time ±10% and ±20% scenarios, selling price remained first, demand stability second, accessibility third, payment speed fourth, and product absorption capacity fifth. No rank reversals occurred, and the Spearman correlation between every perturbed order and the baseline order was 1.000.

Table 3 reports the complete final criterion vector and its robustness range. The weights are ordinal rank-order centroid estimates and are not presented as reconstructed AHP pairwise-comparison results.

4.3 Priority strategies for coconut agroindustry development in Buton Regency

This section reports the group-level SWOT-AHP priorities and their translation into four TOWS strategy orientations. The analysis uses the retained consensus group vector and a transparent additive synthesis of the two SWOT groups associated with each strategy.

4.3.1 Structure of coconut agroindustry development strategy

The group-level synthesis and final strategy weights are shown in Figure 4.

Figure 4 shows how each strategy combines one internal and one external SWOT group. SO combines strengths and opportunities; ST combines strengths and threats; WO combines weaknesses and opportunities; and WT combines weaknesses and threats. This group-level specification is deliberately narrower than a factor-by-factor hierarchy and corresponds to the numerical information retained in the study record.

Strengths represented resource availability, value-added potential, and existing market capacity. Opportunities captured expanding demand, product diversification, and institutional support. Weaknesses reflected technological, productivity, infrastructure, and adoption constraints, while threats represented price volatility, market uncertainty, and climate-related exposure. The value-added and marketing-channel results informed the substantive interpretation of these groups.

The equivalent reciprocal matrix in Table 4 reproduces the retained group vector exactly. Its consistency statistics are λmax=4.0000, CI=0.0000, RI=0.90, and CR=0.0000. These values describe the equivalent ratio matrix and should not be interpreted as reconstructed diagnostics for the unavailable individual expert matrices.

Table 4. Equivalent reciprocal matrix reproducing the retained SWOT-group priority vector.

4.3.2 Priority of agroindustry development strategies

The aggregate weights of the four SWOT groups are presented in Table 5. The equivalent reciprocal matrix is reported in Table 4, and final strategy priorities with sensitivity ranges are presented in Table 6.

Table 5. Aggregate weights of SWOT groups for coconut-agroindustry development in Buton Regency.

Table 6. Group-level SWOT-AHP strategy priorities and sensitivity results.

The equivalent SWOT-group matrix and the final strategy-sensitivity results are presented in Tables 4, 6.

The equivalent matrix was calculated directly as a*(i,j) = w(i)/w(j). Strategy scores were calculated from the normalized sums of their associated SWOT-group weights. All 16 SWOT sensitivity scenarios preserved the baseline strategy order, with zero rank reversals and Spearman ρ = 1.000.

The results presented in Table 5 show that the “Strengths” factor obtained the highest strategic priority weight (0.4658), followed by “Opportunities” (0.2771), “Weaknesses” (0.1611), and “Threats” (0.0960). These findings indicate that coconut agroindustry development in Buton Regency is primarily driven by internal resource capacity, agroindustrial potential, and favorable production conditions. Previous coconut marketing studies similarly indicate that strong farmer participation, existing marketing structures, and substantial coconut resource availability provide important foundations for agroindustry expansion (Rauf et al., 2023).

The relatively high weight assigned to opportunities further suggests that external market conditions and institutional environments provide favorable prospects for future agroindustry development. Expanding demand for coconut-derived products, processing diversification opportunities, and institutional strengthening initiatives collectively support future agroindustrial growth. SWOT-AHP studies similarly emphasize that external opportunities such as market expansion and value-added diversification play critical roles in determining strategic development priorities (Agus et al., 2025; Basar, 2024; Elfahmi et al., 2024; Ismail et al., 2026; Nugraha et al., 2022; Phadermrod et al., 2019).

Combining the group weights produced normalized strategy priorities of 0.3715 for SO, 0.2809 for ST, 0.2191 for WO, and 0.1286 for WT. The SO orientation was therefore dominant, emphasizing value-added processing, institutional strengthening, technology dissemination, and market expansion. ST ranked second because the large strength weight also supports strategies that use existing capacity to manage price, market, and climate threats.

From a behavioral-economics perspective, the SO preference is consistent with stakeholders emphasizing achievable gains when institutional support and market opportunities appear credible. This interpretation remains conceptual because the study did not administer validated measures of optimism, risk aversion, trust, or prospect-theory parameters.

5 Discussion and implications

5.1 Discussion

The findings demonstrate that coconut agroindustry actors in rural market environments do not make decisions solely based on neoclassical assumptions of profit maximization. Instead, marketing institution preferences were strongly shaped by institutional trust, liquidity certainty, transaction security, and accessibility considerations. Agroindustry actors consistently preferred institutions capable of providing faster payment systems, stable product demand, and reliable transactions even when alternative channels potentially offered higher nominal prices. This behavioral pattern reflects bounded rationality, where economic actors simplify complex decisions by relying on familiar institutional arrangements and socially embedded transaction systems under conditions of uncertainty and limited information (Githinji et al., 2023; Křečková and Brožová, 2017; Manevska-Tasevska and Hansson, 2011; Wuepper et al., 2023). Similar findings in Indonesian agricultural marketing systems indicate that institutional trust, patron–client relationships, and information asymmetry strongly influence marketing participation and strategic behavior among rural producers (Chatra and Rosi, 2023; Costanigro et al., 2011; Lestari and Premono, 2019). Consequently, institutional certainty functions as an adaptive mechanism that reduces perceived market risk and supports stable economic participation within imperfect rural markets.

The study also reveals that marketing structure significantly influences both economic outcomes and behavioral responses among coconut agroindustry actors. Shorter marketing channels generated higher farmer's shares and improved marketing efficiency, yet many producers continued to depend on intermediary actors despite lower relative returns. This finding suggests the existence of behavioral lock-in effects and institutional dependency within rural market systems. Intermediaries remained important because they provided essential non-price benefits such as immediate cash payments, informal credit access, transaction flexibility, information support, and market connectivity, mirroring patron-client structures that can stabilize short-term livelihoods while reinforcing longer-term dependency (Ferdous and Nunan, 2025; Ferrol-Schulte et al., 2014; Noak, 2024). Previous studies similarly demonstrate that intermediaries function not only as economic actors but also as institutional bridges linking producers with broader market systems under conditions of limited information and weak infrastructure (Chatra and Rosi, 2023; Lestari and Premono, 2019; Rauf et al., 2023). Therefore, improving marketing efficiency in coconut agroindustry systems requires more than structural market reforms because market behavior is deeply embedded within trust-based institutional relationships and governance structures.

Another important finding concerns the behavioral implications of value-added diversification and strategic agroindustry development. Coconut shell charcoal processing generated substantial value-added and profitability, yet adoption remained geographically limited despite its strong economic potential and the broader sustainability promise of coconut-shell and coconut-husk valorization (D’Almeida and de Albuquerque, 2025; Gokulsidhardh et al., 2026; Singh et al., 2024). This condition indicates that agroindustry innovation is constrained not only by technical and financial limitations but also by behavioral barriers such as risk aversion, operational uncertainty, and low familiarity with new processing systems. The SWOT-AHP results further showed that agroindustry actors prioritized aggressive Strength–Opportunity (SO) strategies emphasizing market opportunities, institutional support, and value-added expansion rather than defensive or risk-minimization strategies (Basar, 2024; Bruzzese et al., 2023; Solangi et al., 2019). This strategic orientation is consistent with prospect theory, which suggests that economic actors become more willing to pursue expansion-oriented behavior when opportunities are framed as achievable gains supported by institutional certainty (Costanigro et al., 2011; Phadermrod et al., 2019; Saaty, 2008). The findings therefore contribute to behavioral economics literature by demonstrating that naturally occurring rural agroindustry systems provide important empirical contexts for understanding strategic adaptation, institutional trust, bounded rationality, and innovation behavior outside laboratory environments.

The integrated framework also clarifies how the economic and decision-analysis components relate. The 54.87% average value-added ratio provides direct evidence for the attractiveness of waste-based diversification, while the marketing-channel comparison demonstrates the distributive advantage of shorter chains. These findings were not mechanically converted into preference or SWOT weights; instead, they informed criterion selection, SWOT-group interpretation, and the substantive reading of the final priorities. The revised manuscript therefore distinguishes empirical measurement from expert-informed synthesis.

The sensitivity analysis demonstrated that the principal policy conclusion was not dependent on small changes in the reported weights. All criterion ranks and all four strategy ranks remained unchanged under ±10% and ±20% one-at-a-time perturbations. The minimum and maximum SO scores were 0.3582 and 0.3838, respectively, and every perturbed strategy order had Spearman ρ = 1.000 relative to the baseline. These results provide deterministic internal validation of local rank stability and indicate that the SO recommendation is robust within the tested range. At the same time, robustness in MCDM should not be equated with a single perturbation exercise. Validation can also involve consistency diagnostics, cross-method triangulation, expert or benchmark comparison, statistical agreement measures, and probabilistic simulation such as Monte Carlo analysis (Sarkar et al., 2026b). Because the original pairwise judgments are unavailable, the present study does not claim these broader forms of validation; instead, it reports the strongest reproducible robustness evidence supported by the retained data.

5.2 Implications

The findings carry important policy implications for coconut agroindustry development and rural institutional strengthening. Policies aimed at improving agroindustry performance should not focus exclusively on production expansion or technical efficiency because marketing behavior and strategic decision-making are strongly influenced by institutional trust, payment certainty, market accessibility, and transaction security (Adawiah et al., 2025). Strengthening farmer cooperatives, improving transparent payment systems, expanding market information access, and reducing institutional uncertainty may therefore be more effective than purely price-based interventions in improving marketing efficiency and reducing excessive dependency on intermediaries (Bheekharry, 2021; Flint, 2004; Folinas and Fotiadis, 2017; Išoraitė, 2024; Madhani, 2016). Previous studies similarly emphasize that governance reform, transparency mechanisms, and improved information flows are essential for strengthening bargaining power and market participation within rural agricultural systems (Lestari and Premono, 2019; Modekurti, 2016; Rauf et al., 2023).

The study also highlights the importance of behaviorally informed agroindustry development strategies capable of addressing cognitive and institutional barriers to innovation adoption. Although coconut shell charcoal processing demonstrated substantial economic potential, adoption remained constrained by perceived operational complexity and uncertainty. This finding suggests that agroindustry policies should incorporate behavioral facilitation mechanisms such as demonstration-based learning, peer-group mentoring, gradual technology adoption schemes, and institutional learning support systems. Behavioral interventions designed to reduce perceived risk and strengthen producer confidence may accelerate agroindustry diversification more effectively than technology dissemination programs alone (Brock and Barham, 2013; Cordaro and Desdoigts, 2021; Liu et al., 2025; Mastenbroek, 2025). Methodologically, the integration of field-based behavioral analysis, multicriteria decision-making, and SWOT-AHP strategic evaluation also demonstrates the importance of moving beyond laboratory-centered behavioral economics approaches toward institutional field contexts where economic behavior is shaped by complex interactions among governance structures, social relationships, market uncertainty, and strategic adaptation processes (Costanigro et al., 2011; Nugraha et al., 2022; Phadermrod et al., 2019).

6 Limitations

The study is geographically limited to seven coconut-producing districts in Buton Regency. The findings may not generalize to regions with different commodity structures, infrastructure, institutional arrangements, or trader–producer relationships. The cross-sectional design captures preferences at one point in time and cannot establish whether institutional trust, liquidity constraints, or channel choices change as prices, policies, and market access evolve. Because prices and institutional conditions were not experimentally manipulated, causal effects cannot be inferred.

The behavioral constructs were not measured using validated psychometric instruments. The preference criteria and observed market choices therefore function as behavioral proxies, and the theoretical contribution should be understood as conceptual and preference-based rather than as direct empirical validation of bounded rationality, institutional trust, risk aversion, or prospect theory. Expert judgments may also be affected by social desirability, professional background, anchoring, and group-composition effects. The ten-expert panel provides contextual knowledge but does not remove subjectivity. Finally, the analysis concentrates on copra and coconut-shell charcoal and does not represent the full range of coconut downstream products.

The expert-panel record retained for this revision contained the final SWOT-group vector and the ordinal marketing-criteria order, but not the ten individual AHP pairwise-comparison matrices. The absence of these original matrices is a direct limitation on the full reproducibility of the research: an independent researcher cannot reproduce the original elicitation process, verify expert-specific consistency ratios, quantify disagreement among experts, or reconstruct the original aggregation procedure from the archived record. The equivalent SWOT matrix reported in this manuscript is only a transparent mathematical representation that reproduces the retained group vector; it is not evidence that the unavailable individual judgments had the same consistency properties. Likewise, the marketing-criteria weights are explicitly identified as ROC estimates derived from the retained ordinal ranking rather than reconstructed AHP weights. These constraints limit reproducibility of the original weighting process even though the revised calculations based on the retained vectors are fully transparent and reproducible. Future data-management protocols should archive all individual pairwise matrices, expert identifiers in de-identified form, aggregation rules, analysis scripts, and diagnostic outputs. For the same reason, the numerical margins by which the ROC-derived marketing criteria are separated (Section 4.2.3, Marketing-Institution Preference Criteria) should be read as descriptive of the retained ordinal consensus rather than as reconstructed measures of expert preference intensity.

7 Future research

Future research should compare coconut-producing regions with different institutional and market structures, use longitudinal designs to observe preference changes over time, and combine field experiments with validated behavioral scales. Experimental variation in payment timing, price transparency, cooperative access, and purchase guarantees would help identify causal mechanisms. Further studies should compare ROC, AHP, FUCOM, LBWA, and DIBR weighting results on the same fully archived expert dataset and evaluate convergence through cross-method triangulation. Robustness analysis should also extend beyond deterministic one-at-a-time perturbations by incorporating scenario-based tests, Monte Carlo or other probabilistic simulations, expert validation, and statistical rank-agreement measures where the data structure permits (Sarkar et al., 2026b). Broader product coverage should include coconut oil, sugar, fiber, beverages, and other circular-economy pathways.

8 Conclusion

This study examined how economic outcomes, marketing structures, and institutional preferences shape coconut-agroindustry development in Buton Regency. Coconut-shell charcoal processing generated an average value added of IDR 1,002.52/kg and an average value-added ratio of 54.87%, demonstrating meaningful diversification potential. The shorter marketing channel produced a lower margin and a higher farmer's share than the intermediary-intensive channel, although producers continued to value the liquidity, access, and transaction security provided by intermediaries.

The preference analysis identified selling price (0.4567), demand stability (0.2567), and accessibility (0.1567) as the leading criteria. The SWOT-group priorities favored strengths (0.4658) and opportunities (0.2771), producing a robust strategy order of SO (0.3715), ST (0.2809), WO (0.2191), and WT (0.1286). These findings support a development orientation centered on value-added processing, institutional strengthening, reliable payment systems, market information, technology dissemination, and market expansion. The study achieves its practical objective by connecting measurable economic evidence with transparent preference and strategy synthesis. Its behavioral contribution is appropriately bounded: market choices are interpreted in relation to uncertainty, liquidity, transaction security, and institutional familiarity, rather than as direct psychometric validation of behavioral constructs.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Darko Bozanic, University of Defense, Serbia

Reviewed by: Adis Puška, University of Bijeljina, Bosnia and Herzegovina

Dragan Pamucar, University of Belgrade, Serbia