{"stable_id":"968b26cd2d0b5f01","slug":"cottonace-pest-advisory","company":"Wadhwani AI CottonAce partner programs","workflow_name":"AI-counted pink-bollworm traps and threshold-based spray advisories","function_code":"agriculture","pattern_codes":["threshold_as_control"],"changed_assumption":"Farmers can spray against a shared trap-count threshold instead of a calendar or subjective visual assessment.","evidence_strength":"verified","publication_tier":"showcase","freshness":"watch","reviewed_at":"2026-08-23","updated_at":"2026-08-22","source_quality_summary":"1 independent, 1 peer reviewed; publication outcomes are verified.","caveat_summary":"The mixed evidence is essential: early evaluations were not all randomized, and unusually high rainfall with low pest pressure eliminated significant benefit in the later multi-state experiment.","freshness_caveat":"Watch status: verify the cited source and deployment condition before reusing this case.","workflow_summary":{"before":[{"actor":"Cotton farmer","action":"Inspect fields and decide pesticide timing from experience, visible damage, or calendar practice.","actor_type":"control"},{"actor":"Farmer","action":"Apply pesticide, sometimes before or after the economically effective pest threshold.","actor_type":"control"}],"hinge":"Farmers can spray against a shared trap-count threshold instead of a calendar or subjective visual assessment.","after":[{"actor":"Lead farmer","action":"Checks pheromone traps weekly and uploads smartphone images.","actor_type":"system"},{"actor":"CottonAce","action":"Detects and counts pink bollworm moths and compares counts with an action threshold.","actor_type":"system"},{"actor":"Farmers","action":"Receive a localized spray advisory and decide whether and how to apply treatment.","actor_type":"system"}],"decision_mode":"shared","decision_marker":"Human sets the threshold; AI decides each instance"},"before":[{"order":1,"actor":"Cotton farmer","action":"Inspect fields and decide pesticide timing from experience, visible damage, or calendar practice.","handoff_to":"Spray operation","control":"Farmer judgment"},{"order":2,"actor":"Farmer","action":"Apply pesticide, sometimes before or after the economically effective pest threshold.","handoff_to":"Crop","control":"Available pesticide and labor"}],"after":[{"order":1,"actor":"Lead farmer","action":"Checks pheromone traps weekly and uploads smartphone images.","handoff_to":"CottonAce model","control":"Manual trap placement and image capture"},{"order":2,"actor":"CottonAce","action":"Detects and counts pink bollworm moths and compares counts with an action threshold.","handoff_to":"Farmer group","control":"Integrated-pest-management threshold"},{"order":3,"actor":"Farmers","action":"Receive a localized spray advisory and decide whether and how to apply treatment.","handoff_to":"Crop","control":"Farmer retains application authority"}],"decision_rights":"The model determines whether the observed count crosses an advisory threshold; farmers retain the pesticide application decision.","exception_path":"Farmers and extension partners can withhold treatment, inspect manually, or use other integrated-pest-management methods; rainfall and low pest pressure can make the advisory produce no measurable benefit.","removed_work":["Subjective manual moth counting for every participating farmer","Some prophylactic or mistimed pesticide applications"],"outcomes":[{"metric":"Farmer income in initial field evaluations","baseline":"Non-adopter or control farmers in 2020-2021 evaluation locations","result":"Up to 22% higher income in the first-year evaluations; no significant benefit in the 2021-2022 multi-state experiment","period":"2020-2022","scale":"Initial evaluations in Ranebennur and Wardha followed by a seven-state experiment","attribution_caveat":"The mixed evidence is essential: early evaluations were not all randomized, and unusually high rainfall with low pest pressure eliminated significant benefit in the later multi-state experiment.","evidence_label":"verified"}],"executive_lesson":"This case is valuable because the larger later test was null: AI advice creates value only when the targeted risk is present and the human sensing routine is sustainable.","anti_pattern":"Promoting the 22% first-year result without the later null result or ignoring the labor required to inspect traps.","questions_for_leaders":["What external condition determines whether the intervention can help?","Is the human data-collection burden sustainable?"],"collections":[],"bundle_version":"1.0.0","bundle_fingerprint":"sha256:c23c6cc2b88153d008ea8fda928f632ce0011fc2d4c5036672a16e5d895bab93","canonical_url":"https://brianletort.ai/transformations/cottonace-pest-advisory"}