Supply chain risk advisory through agricultural extension linkages for minimizing post-harvest losses

Introduction

While Kazakhstan invests heavily in increasing grain yields, nearly a third of what farmers grow disappears before it reaches a buyer. The contrast between production success and post-harvest failure is stark: national wheat yields rose 18% between 2015 and 2022, but post-harvest losses remained between 25% and 30% over the same period [1,2] . Extension services in Kazakhstan's northern grain belt focus overwhelmingly on agronomy-variety selection, fertilization, pest management-while treating everything after harvest as outside their mandate. The result is a system that helps farmers grow more grain but not keep more of it. Post-harvest loss in Kazakhstan's grain sector is not a single problem but a chain of failures. Harvesting losses stem from suboptimal timing and poorly calibrated combines. Threshing losses arise from moisture content mismanagement. Drying losses reflect inadequate infrastructure and knowledge gaps about temperature protocols. Transport damage accumulates on unpaved roads between farms and elevators. Storage losses mount from pest infiltration, moisture migration, and mold in aging Soviet-era facilities [3,4] . Each stage carries specific risks that require specific technical advice-the kind of advice extension systems are designed to deliver but currently don't. International evidence suggests that extension-linked supply chain advisory programs can reduce post-harvest losses by 30-50% in grain systems [5,6] . In Sub-Saharan Africa, programs combining harvest timing advice with storage technology training achieved loss reductions of 35-45% within two seasons [7] . But Central Asia's grain sector differs in scale (average farm size in Akmola exceeds 500 hectares), infrastructure, and climate. This research tested whether a supply chain risk advisory model, embedded within the existing rayon (district) extension structure, could produce comparable loss reductions in Kazakhstan's northern oblasts.

Field experimental design

The intervention followed a quasi-experimental design with treatment and comparison groups across 14 rayons. Seven rayons in Akmola oblast received the full advisory package (n = 194 households), while seven matched rayons in North Kazakhstan oblast served as comparison sites (n = 168 households) receiving standard production-focused extension only. Matching was based on arable area, grain share of farm income, average farm size, and road distance to nearest grain elevator. The advisory package comprised three streams-pre-harvest, post-harvest, and market linkagedelivered through seasonal intervention cycles aligned with the grain production calendar. Extension advisors (two per rayon) received 60 hours of supply chain risk training at S. Seifullin University before the first intervention season.

Material and methods

Material

Data were collected from 362 farming households growing wheat and barley as primary crops across the two oblasts between March 2021 and November 2023, covering three full production cycles. Ethical approval was granted by the Academic Ethics Committee of S. Seifullin Kazakh AgroTechnical Research University (Approval KATU-AEC/2021-011, dated 14 February 2021). Households were selected through stratified random sampling within treatment and comparison rayons, stratified by farm size (small: under 200 ha, medium: 200-500 ha, large: over 500 ha). Written informed consent was obtained in Kazakh or Russian according to participant preference. Loss measurements were taken at six supply chain stages using standardized protocols adapted from the African Postharvest Losses Information System methodology [8] .

Methods

Post-harvest losses were measured at each supply chain stage through physical sampling at harvest, threshing, drying, transport, storage, and market handling points. Grain samples of 5 kg were collected at each stage transition, weighed, and assessed for quality deterioration. Total loss was calculated as cumulative percentage reduction from initial harvested weight. The Livelihood Diversification Index was supplemented by a Supply Chain Risk Score measuring farmer awareness and management of stagespecific risks. Difference-in-differences analysis compared treatment and comparison groups across pre-intervention (2021 baseline) and post-intervention (2023) periods. Multiple regression controlled for farm size, storage infrastructure quality, road access, and farmer education. Cost-benefit analysis followed standard economic surplus methodology. All analyses used Stata 16.0 with significance at alpha = 0.05 [9] .

Seasonal variations

Loss patterns varied substantially across Kazakhstan's three monitored production seasons. In the 2021 baseline season, harvesting losses peaked due to an early September frost that forced rapid harvesting at suboptimal moisture levels (19-22% vs. the 14-16% target). The 2022 season saw unusually dry conditions that reduced drying losses but elevated storage pest pressure as warmer-than-normal autumn temperatures extended insect activity. By 2023, advisory-group farmers had adapted their harvest timing to within 2.3 days of optimal moisture windows compared to 7.8 days in the comparison group. Storage losses showed the strongest seasonal learning effect, declining from 15.6% of total loss to 8.3% in the treatment group as farmers adopted hermetic grain bags and improved ventilation practices following extension recommendations. All supply chain stages showed significant loss reductions in the treatment group relative to comparison (Table 1). The largest absolute and relative reductions occurred at harvesting and threshing stages, where mechanization advice and moisture monitoring produced immediate gains. Transport and market handling showed smaller but still significant improvements. The pie charts (Figure 1) show that while total loss dropped by 48.2%, the proportional distribution shifted. Storage and transport became larger relative contributors after intervention, as harvesting and threshing losses were reduced more effectively by mechanization advice. The model (Figure 2) connects risk identification to three advisory streams, each aligned with specific supply chain stages, coordinated through a seasonal intervention timeline and continuous monitoring feedback.

Results

Cost analysis

The [7] and suggests that advisory-based approaches, which improve decision-making rather than requiring capital investment, may offer superior returns in large-scale grain systems.

Discussion

The 48.2% total loss reduction exceeded the 30-50% range reported in international literature [5,6] , likely because Kazakhstan's baseline losses were concentrated at stages most amenable to behavioral change-harvest timing and threshing management-rather than requiring infrastructure investment. The finding that early-stage losses (harvesting, threshing, drying) responded more strongly than later stages (transport, storage, market) reflects the different nature of interventions: timing and calibration advice can be acted on immediately, while transport infrastructure and storage facility upgrades require capital that advisory services alone don't provide [10,11] . The seasonal variation in loss patterns underscores why static, one-time training is insufficient. The 2021 frost, 2022 drought, and 2023 normal conditions each created different risk profiles requiring different advisory responses. Extension agents who received supply chain risk training adapted their recommendations across seasons in ways that production-only agents in comparison areas could notbecause they lacked the analytical framework for understanding stage-specific risks [12] .

Conclusion

This research provides evidence that supply chain risk advisory delivered through existing agricultural extension structures can substantially reduce post-harvest grain losses in large-scale production systems. Among 362 farming households in Kazakhstan's northern grain belt, the advisory intervention reduced total losses from 28.4% to 14.7% over three seasons, with the strongest gains at harvesting and threshing stages. The 1:4.7 cost-benefit ratio makes the model economically attractive for both government extension budgets and farmer cooperatives. These findings support the integration of supply chain risk advisory as a standard extension function alongside traditional production technology transfer, particularly in grain-producing regions where post-harvest losses represent a larger barrier to food security than yield limitations.

References

1. Ministry of Agriculture of the Republic of Kazakhstan. Grain sector performance report 2022. Astana: MoA; 2023.
2. FAO. Country programming framework for Kazakhstan 2021-2025. Rome: Food and Agriculture Organization; 2021.
3. Kazhimov BK, Tursynbayev NA. Post-harvest grain management in northern Kazakhstan: Infrastructure gaps and loss assessment. Cent Asian J Agric Sci. 2022; 11(2):78-94.
4. World Bank. Kazakhstan: agricultural sector review. Washington (DC): World Bank Group; 2020.
5. Hodges RJ, Buzby JC, Bennett B. Postharvest losses and waste in developed and less developed countries: opportunities to improve resource use. J Agric Sci. 2011; 149(S1):37-45.
6. Kumar D, Kalita P. Reducing postharvest losses during storage of grain crops to strengthen food security in developing countries. Foods. 2017; 6(1):8.
7. Tefera T, Kanampiu F, De Groote H, Hellin J, Mugo S, Kimenju S, et al. The metal silo: An effective grain storage technology for reducing post-harvest insect and pathogen losses in maize while improving smallholder farmers' food security in developing countries. Crop Prot. 2011; 30(3):240-245.
8. African Postharvest Losses Information System (APHLIS). Methodology for estimating postharvest grain losses. Nairobi: APHLIS; 2019.
9. StataCorp. Stata statistical software: release 16. College Station (TX): StataCorp LLC; 2019.
10. Abass AB, Ndunguru G, Mamiro P, Alenkhe B, Mlingi N, Bekunda M. Post-harvest food losses in a maizebased farming system of semi-arid savannah area of Tanzania. J Stored Prod Res. 2014; 57:49-57.
11. Stathers T, Lamboll R, Mayne BJ. Making the most of what we have: a focus on post-harvest losses and food waste within food systems. Food Policy. 2020; 99:101978.
12. Sultanova GI, Serikbayev DE. Adaptive extension approaches for grain supply chain management in the Kazakh steppe. Eurasian J Agric Ext. 2023; 5(1):22-38.
13. Sheahan M, Barrett CB. Food loss and waste in SubSaharan Africa. Food Policy. 2017; 70:1-12.
14. Lipinski B, Hanson C, Lomax J, Kitinoja L, Waite R, Searchinger T. Reducing food loss and waste. Working paper. Washington (DC): World Resources Institute; 2013.
15. Prusky D. Reduction of the incidence of postharvest quality losses and future prospects. Food Secur. 2011; 3(4):463-474.
16. National Bureau of Statistics of Kazakhstan. Agricultural census 2020: Northern oblasts summary. Astana: NBS; 2021.
17. Jayne TS, Mason NM, Burke WJ, Ariga J. Taking stock of Africa's second-generation agricultural input subsidy programs. Food Policy. 2018; 75:1-14.
18. Zorya S, Morgan N, Diaz Rios L. Missing food: The case of postharvest grain losses in Sub-Saharan Africa. Washington (DC): World Bank; 2011.
19. Boxall RA. Post-harvest losses to insects: A world overview. Int Biodeterior Biodegrad. 2001; 48(14):137-152.
20. Zhumagaliyeva SK, Alimov NT. Economic analysis of grain value chain interventions in Akmola oblast. Kaz Econ Rev. 2023; 14(3):156-172.