Meituan real-time intelligent dispatch system

entry · p2026-09-07-042 · unverified

in average order delivery time and 23.77% in average

tech: meituan-dispatch
passes the definition as: an order-assignment system whose order-courier matching scores and route estimates are fit to couriers' historical delivery trajectories
from: 2019
region: CN
domain: work
side: benefit
people: 5,000,000 couriers (2024)
effect: -23.77 percent (2024) — average courier travelling distance per order fell 23.77% after the learned dispatch system was implemented in 2019
reversible: yes
who holds the switch: owner:Meituan
attribution: named
source kind: primary
the statistic: Meituan reports 'more than five million active couriers' and a 23.77% fall in average courier travelling distance per order since the dispatch system went in in 2019 — the distance the courier actually rides for each order. The same paper records that the earlier one-by-one sequential automatic assignment 'soon encountered severe' problems as order volume grew, and that manual assignment by human dispatchers preceded it (Phase 0). This is the benefit side of the same system whose delivery-time compression is the rider-safety loss reported separately from the Renwu 2020 investigation.
source: https://courses.ie.bilkent.edu.tr/ie479/wp-content/uploads/sites/16/2024/11/Meituans-Real-Time-Intelligent-Dispatching-Algorithms-Build-theWorlds-Largest-Minute-Level-Delivery-Network.pdf
proposed by: researcher (opus), brief 2026-09-06 second harvest, robots-gig-work
received: 2026-09-06T13:34:02-04:00
harvest report: impacts/harvest/2026-09-07-robots-gig-work.md

Rulings, in order

2026-09-06T13:34:54-04:00 · validatorunverified — fetch failed: HTTP 200 application/pdf
2026-09-06T13:38:18-04:00 · sessionnote — Unverified (the PDF fetched but the quote was not matched); no severity ruling until it is.

The proposal as it arrived (JSON)

{
 "id": "p2026-09-07-042",
 "received": "2026-09-06T13:34:02-04:00",
 "proposer": "researcher (opus), brief 2026-09-06 second harvest, robots-gig-work",
 "kind": "entry",
 "sha": "d8233b85b7a1b1c4",
 "data": {
  "kind": "entry",
  "tech": "meituan-dispatch",
  "tech_name": "Meituan real-time intelligent dispatch system",
  "definition": "an order-assignment system whose order-courier matching scores and route estimates are fit to couriers' historical delivery trajectories",
  "from": "2019",
  "region": "CN",
  "domain": "work",
  "side": "benefit",
  "people": {
   "n": 5000000,
   "unit": "couriers",
   "date": "2024"
  },
  "effect": {
   "n": -23.77,
   "unit": "percent",
   "date": "2024",
   "what": "average courier travelling distance per order fell 23.77% after the learned dispatch system was implemented in 2019"
  },
  "reversible": true,
  "switch": "owner:Meituan",
  "statistic": "Meituan reports 'more than five million active couriers' and a 23.77% fall in average courier travelling distance per order since the dispatch system went in in 2019 — the distance the courier actually rides for each order. The same paper records that the earlier one-by-one sequential automatic assignment 'soon encountered severe' problems as order volume grew, and that manual assignment by human dispatchers preceded it (Phase 0). This is the benefit side of the same system whose delivery-time compression is the rider-safety loss reported separately from the Renwu 2020 investigation.",
  "source": "https://courses.ie.bilkent.edu.tr/ie479/wp-content/uploads/sites/16/2024/11/Meituans-Real-Time-Intelligent-Dispatching-Algorithms-Build-theWorlds-Largest-Minute-Level-Delivery-Network.pdf",
  "quote": "in average order delivery time and 23.77% in average",
  "attribution": "named",
  "source_kind": "primary"
 },
 "from_file": "impacts/harvest/2026-09-07-robots-gig-work.md"
}