{"id":"359b3545-341c-4fcd-ac41-38ae4729afe3","arxiv_id":"2606.29248","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.5,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Supply-chain-aware XGBoost–LightGBM ensembles predict Sri Lankan vegetable prices with ~90.8% holdout accuracy and retain ~86% accuracy on unseen 2024 hyperinflation without retraining.","lead":"A gradient-boosted model with supply-chain features predicts Sri Lankan vegetable prices at about 91% accuracy (1−MAPE) and still tracks major 2024 hyperinflation surges without retraining. It matters because import-isolated food markets have almost no external buffer, so early warning of week-scale price spikes can protect farmers, traders, and consumers.","discovery_kind":"new_application","skeptic_critique":{"model":"grok-4.5","headline":"Cross-regime claim rests on untested dual-origin imputation and market scoping that may not recover true 2024 supply corridors.","rationale":"The reader correctly isolates the dual-origin imputation and scoping as the weakest assumption underwriting both the strongest predictor (lagged spread) and the headline cross-regime result. No internal contradiction appears; the empirical pipeline is coherent once those features are accepted. The concern is therefore not fatal but load-bearing: without a sensitivity check the 85.96% transfer and “structural mechanics” interpretation remain conditional on untested preprocessing. That matches the reader’s CONDITIONAL verdict and the call for imputation/selection stress-tests. No stronger independent flaw (e.g., leakage or non-reproducible math) displaces this one. Fixing CI reporting and releasing code remain necessary but secondary to settling whether the supply-chain features are faithful.","tokens_in":9075,"tokens_out":583,"duration_ms":5943,"concrete_test":"Re-impute farmer-gate prices under three alternatives (single nearest historical origin; leave-one-of-top-two-out; random supplier from the same agro-climatic zone) and re-map weather origins with a 2020–2023 production-share update if available; retrain the identical Optuna-tuned ensemble and re-evaluate 2018–2019 holdout plus the 2024 Green Chillies Kaluthara surge (Table V). If 2024 accuracy falls below ~80% or Week-29 error exceeds ~5% under any plausible alternative, the structural-transfer claim weakens.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim that supply-chain mechanics make import-isolated prices structurally predictable (and transferable to 2024 hyperinflation without retraining) depends on Strategies 2–3 (§III) correctly recovering the true farmer-gate and origin-weather state. Dual-origin arithmetic-mean imputation of urban farmer-gate prices (from historically top-two suppliers) and geographic-mean weather from mapped cultivation districts are treated as ground truth; the lagged farmer-to-retail spread then becomes the strongest predictor and the 85.96% 2024 accuracy (Table IV, Green Chillies micro-analysis) is offered as evidence of structural transfer. If the 2013–2019 historical supplier ranking or the “most consistent” 12-variety/14-market scope systematically misrepresents logistics corridors that shifted under crisis (fuel shortages, route changes, different production zones), both the in-sample importance of the spread and the out-of-time accuracy would be overstated. The paper reports no sensitivity to alternative origin maps, leave-one-supplier-out imputation, or re-scoping, so the load-bearing assumption remains untested.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The paper constructs an integrated Sri Lankan vegetable-price panel (12 varieties, 14 markets, 2013–2019) that aligns retail and farmer-gate prices with origin-zone weather lags, diesel costs, and USD/LKR rates, then trains Optuna-tuned XGBoost–LightGBM ensembles in unified and season-specific (Maha/Yala) configurations. On a chronological 2018–2019 holdout the unified ensemble reports R² = 0.9281 and 90.84% accuracy (defined as 1−MAPE); the Yala-only model reaches the highest within-season R² (0.9420). The same 2013–2019 model, without retraining, retains 85.96% accuracy on a fully unseen 2024 hyperinflationary period and tracks a large Green Chillies surge with sub-1% error at the inflection week. The authors conclude that prices in fully import-isolated markets are structurally predictable once supply-chain mechanics are encoded, and that seasonal segmentation trades within-season fit against cross-season momentum.","tokens_in":9411,"tokens_out":857,"duration_ms":6802,"significance":"If the cross-regime result holds under scrutiny, the work supplies a concrete, falsifiable demonstration that supply-chain-aware tabular ensembles can transfer across a fivefold inflation regime in an import-isolated agricultural market—something prior ARIMA/GARCH studies of Sri Lankan vegetables never attempted. The public data-and-code commitment, expanding-window CV with reported CIs, ablation on diesel/FX, and micro-analysis of a genuine supply shock are genuine strengths that raise the bar for applied agricultural forecasting. The practical payoff (early-warning value for farmers, traders, and food-security policy) is clear and well-motivated.","major_comments":[{"comment":"§III Strategies 2–3 and the RQ5 claim (Table IV, Green Chillies micro-analysis): the dual-origin arithmetic-mean imputation of urban farmer-gate prices and the geographic-mean weather from historically mapped cultivation districts are treated as ground truth; the lagged farmer-to-retail spread then becomes the strongest predictor and the 85.96% 2024 accuracy is offered as evidence of structural transfer. No sensitivity to alternative origin maps, leave-one-supplier-out imputation, or re-scoping of the 12×14 panel is reported. Because logistics corridors and production zones can shift under crisis (fuel shortages, route changes), this is a load-bearing and currently untested assumption for the central cross-regime claim.","section":null},{"comment":"Tables II–IV and abstract: “accuracy” is defined as 1−MAPE. While convenient, this non-standard metric inflates readability and is not comparable to the literature baselines the paper cites (e.g., Champika & Mugera’s 71% SARIMA figure). The manuscript should report MAPE (or sMAPE/RMSE) as the primary error metric alongside R², and restate the 90.84%/85.96% figures accordingly so that the magnitude of the claimed improvement is transparent.","section":null},{"comment":"§V.A Stage 1 and related-work claims: the only quantitative baseline given is a 71.20% time-series accuracy attributed to prior SARIMA work. No ARIMA/SARIMA/Prophet or pure lag-only gradient-boosted model is re-estimated on the same 12×14 panel and the same chronological holdout. Without that head-to-head comparison it is impossible to isolate how much of the 90.84% figure is attributable to the supply-chain feature set versus the richer multi-market panel itself.","section":null}],"minor_comments":[],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The one thing worth knowing is that they built a usable multi-market, multi-crop Sri Lankan vegetable panel with origin-zone weather, diesel, and farm-gate spreads, then showed a plain XGBoost–LightGBM ensemble holds ~86% 1−MAPE accuracy on completely unseen 2024 hyperinflation without retraining and tracks a real Green Chillies spike to sub-1% error at the inflection. That cross-regime result is the actual contribution.\n\nWhat is new is not the learner—gradient boosting, Optuna, lags, and seasonal splits are standard—but the data construction and the stress test. Prior Sri Lankan work was ARIMA/GARCH on single markets. They scoped to 12 varieties × 14 markets, reverse-engineered dual-origin farm-gate imputation for urban hubs, averaged weather from mapped cultivation districts rather than retail cities, and compared unified vs Maha/Yala models. Chronological holdout, expanding-window CV, ablation on diesel/FX, and the 2024 micro-case are real evidence. The diesel paradox (R² rises slightly when diesel is removed, MAPE worsens) is a useful observation about step-change features. Code and data are stated to be public.\n\nSoft spots, in proportion: “accuracy” as 1−MAPE is nonstandard and flatters the headline; some R² CIs go above 1, which is just bad reporting; classical baselines are thin (one 71% SARIMA number); and the dual-origin imputation plus “most consistent” market scoping are load-bearing and untested by sensitivity. The stress-test note is right that if 2024 logistics corridors shifted, the transfer claim is overstated—but the paper still shows the model tracks absolute levels outside the training range using relative supply-chain signals, which is more than most applied papers attempt. No circularity; free parameters are ordinary learner knobs.\n\nThis is for people who care about food-price early warning in import-isolated settings or who need a concrete template for origin-aligned features. It does not reorganize theory. I would send it to peer review; a serious referee can force CI fixes, baseline tables, and imputation sensitivity without killing the result. Worth engaging if you work on agricultural forecasting or crisis transfer.","headline":"Solid applied forecasting paper: real multi-market panel, origin-aligned features, and unretrained 2024 transfer; soft spots are metric reporting and untested imputation, not the core claim.","tokens_in":10086,"tokens_out":560,"would_cite":true,"duration_ms":5489,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"Vegetable prices in fully import-isolated markets are structurally predictable when models encode supply-chain mechanics, not random walks.","keywords":["agricultural price forecasting","XGBoost","LightGBM","import-isolated markets","seasonal modelling","supply-chain feature engineering","cross-regime generalisation","Sri Lanka"],"falsifier":"Re-run the identical pipeline after replacing dual-origin farmer-gate imputation and origin-averaged weather with pure retail-city weather and no farmer-gate features; if hold-out accuracy collapses below the paper’s SARIMA baseline (~71%) and the 2024 surge-tracking error rises well above a few percent, the claim that supply-chain mechanics are what make prices predictable fails.","tokens_in":9964,"feed_emoji":"🥬","tokens_out":1021,"duration_ms":8190,"temperature":0.7,"pith_summary":"In markets cut off from imports, a flood, drought, or logistics break can double vegetable prices in a week because nothing buffers the shock. This paper argues those jumps are not pure noise: they follow the supply chain. The authors build an integrated 2013–2019 panel for twelve vegetables and fourteen Sri Lankan markets that lines weather up to actual growing zones, recovers farmer-gate prices for urban hubs, and folds in diesel costs, exchange rates, and the two monsoon seasons (Maha and Yala). A gradient-boosted ensemble then forecasts retail prices. Season-specific models fit their own windows more tightly, yet a single unified model posts the best overall accuracy and, without retraining, still tracks major surges through the unseen 2024 hyperinflation. The practical claim is early warning for farmers, traders, and food-security authorities who currently lack any supply-chain-aware seasonal forecast.","feed_headline":"Prices that double in a week are still forecastable","feed_subtitle":"Supply-chain features let a 2013–2019 model track 2024 vegetable surges without retraining","key_machinery":"Supply-chain-aware feature engineering plus a weighted XGBoost–LightGBM ensemble: origin-averaged weather lags (1/4/8 weeks), dual-origin farmer-gate imputation, lagged farmer-to-retail spread, diesel and USD/LKR step features, cyclical week encoding, and Optuna-tuned blending (weights ≈0.52/0.48), trained both unified and season-segmented.","core_discovery":"Agricultural price movements in a fully import-isolated market are structurally predictable once the model encodes origin-zone weather lags, diesel-driven logistics, farmer-gate-to-retail spreads, and the two cultivation seasons. On held-out 2018–2019 data the unified ensemble reaches 90.84% accuracy (1−MAPE) and R² 0.9281; it retains 85.96% accuracy on completely unseen 2024 hyperinflation and records sub-1% error at a sharp Green Chillies surge inflection, showing that relative supply-chain dynamics transfer across economic regimes.","pith_inferences":["The same origin-aligned weather-plus-logistics template should transfer to other import-isolated or island economies with dual monsoon calendars and highland/lowland crop splits.","If lagged farmer-to-retail spread is the strongest single predictor, real-time producer-price reporting becomes a higher-leverage public-data investment than denser retail weather stations.","Inflation-index normalisation of the target (suggested in the paper’s future work) would directly test whether residual 2024 R² drop is pure level extrapolation or genuine concept drift in the supply corridors themselves."],"forward_implications":["Early-warning systems can flag week-ahead retail spikes for farmers, traders, and food-security agencies without waiting for absolute price levels to reappear.","Season choice (unified vs Maha/Yala-specific) should be driven by whether transition-week accuracy or within-season precision is the operational priority.","Macro step-changes such as diesel price jumps matter most for extreme spikes even when they barely move overall R², so evaluation must track MAPE on shocks, not only fit statistics.","Cross-regime transfer of relative supply-chain dynamics implies models need not be retrained after every inflation regime if features remain structural rather than level-based."],"fun_headline_variants":["Supply-chain ML forecasts vegetable prices that double in a week","Season-aware models hit 91% accuracy on isolated market volatility","2013-2019 ensemble tracks 2024 surges without retraining","Encoding seasons and logistics makes farm price spikes predictable","Unified model retains 86% accuracy on unseen hyperinflation data"],"cache_read_input_tokens":128,"weakest_assumption_plain":"That averaging historical supplier markets for missing urban farmer-gate prices and averaging weather from mapped cultivation districts recovers the true logistics state that actually drives retail prices, so patterns learned on the reduced 2013–2019 panel transfer to 2024.","fun_headline_variants_meta":{"raw":{"variants":["Supply-chain ML forecasts vegetable prices that double in a week","Season-aware models hit 91% accuracy on isolated market volatility","2013-2019 ensemble tracks 2024 surges without retraining","Encoding seasons and logistics makes farm price spikes predictable","Unified model retains 86% accuracy on unseen hyperinflation data"]},"model":"grok-4.5","effort":"low","cost_usd":0.005568,"raw_usage":{"total_tokens":1626,"prompt_tokens":951,"num_sources_used":0,"completion_tokens":70,"cost_in_usd_ticks":55680000,"prompt_tokens_details":{"text_tokens":951,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":605,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":951,"tokens_out":70,"duration_ms":5198,"temperature":1.0,"reasoning_tokens":605,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-12T11:02:10.424734+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Re-run the identical pipeline after replacing dual-origin farmer-gate imputation and origin-averaged weather with pure retail-city weather and no farmer-gate features; if hold-out accuracy collapses below the paper’s SARIMA baseline (~71%) and the 2024 surge-tracking error rises well above a few percent, the claim that supply-chain mechanics are what make prices predictable fails.","supporting_citations":[],"review_version":2}