Kunming incinerators keep dioxin far below target with AI monitoring

GEF's evaluation office found that AI monitoring kept dioxin at two Kunming incinerators far below target, as low as 0.0046 ng TEQ/m³. World Bank reviewers name no AI.

  • 2 society

    Dioxin emissions at both incinerators stayed below the 0.1 ng TEQ/m³ target, with Xishan as low as 0.0046 ng TEQ/m³, and a third party verified the levels.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Reported benefit.
Evidence

foundstudiedreportedclaimed

An independent evaluation report supports the benefit.

Compared with

The 0.1 ng TEQ/m³ dioxin target. Measured.

Without the system
0.1 ng TEQ/m³ target
With it
As low as 0.0046 ng TEQ/m³ at Xishan
The system’s part
Not isolated. The GEF office credits artificial intelligence with automating combustion monitoring but does not describe the model or separate its share of the dioxin result. The World Bank reviewers of the same project describe a SCADA system for real-time monitoring of combustion and do not mention artificial intelligence. They credit the dioxin results to 35 renovations at Konggang and 16 at Xishan. Neither document shows that the system learns from data.
Limits
Not yet checked.
Costs and harms
Not yet checked.
Later
Expanded, 19 November 2025. The GEF office reports that the system was replicated in 22 incinerators across 13 provinces. In 2022 the World Bank's reviewers described this as a parent company's announced plan to copy the renovation approach.

Download data Cite Suggest a correction

Benefit type
Improved Environmental Sustainability
System
Combustion monitoring automated with artificial intelligence at the Xishan and Konggang incinerators (a SCADA system in the World Bank's review)
Made by
Not named in the evaluations
Used by
Operators of the Xishan and Konggang incinerators in Kunming, under the GEF-financed China Municipal Solid Waste Management Project implemented by the World Bank
Where
Kunming, Yunnan, China
Stage
In routine use
Date
Kind
Benefit created
Broad benefit
Monitoring Processes, Materials, or Surroundings, Analyzing Data or Information
Collections
Scrutiny and findings
Last revised
31 October 2026

Sources

  1. Evaluation report, by authors with no tie to the system.

    Global Environment Facility Independent Evaluation Office. Evaluation of GEF Support for Technological Innovation. GEF/E/C.70/Inf.02, prepared by the Independent Evaluation Office for the 70th GEF Council Meeting, 19 November 2025. A field mission of the office visited the project.

    The Municipal Solid Waste Management Project (GEF ID 4617, World Bank) introduced artificial intelligence to automate combustion monitoring at the Xishan and Konggang incinerators, maintaining dioxin emissions well below Stockholm Convention standards—with Xishan achieving levels as low as 0.0046 ng TEQ/m³, far surpassing the 0.1 ng TEQ/m³ target.

  2. Evaluation report, by authors with no tie to the system. Context or contrary evidence.

    World Bank Independent Evaluation Group. Implementation Completion Report (ICR) Review: CH GEF Municipal Solid Waste Management (P126832), ICRR0022775, 14 February 2022. In the review, the Independent Evaluation Group confirms the dioxin results, describes a SCADA system and does not mention artificial intelligence.

    A new state-of-the-art Supervisory Control and Data Acquisition (SCADA) system was established which gathered and analyzed real-time data to monitor the combustion process of the Kunming and Ningbo incinerators.

Details

AI system

Workflow

The project introduced artificial intelligence to automate combustion monitoring at the two incinerators. Audits of the incinerators led to improvement programs, and the plants carried out 35 and 16 renovations with co-financing. An accredited third party monitored and verified dioxin levels at project closing.

Follow-up

Conditions for replication

Incinerators and regulators must share real-time operating and emission data. The plants must apply the best available techniques and best environmental practices of the Stockholm Convention. A certified laboratory must sample dioxin levels on a regular schedule.

Entry details

Sector

Environment and climate.

Purpose

An unnamed developer built the monitoring system to automate combustion monitoring at the Xishan and Konggang incinerators and help keep dioxin emissions down.

Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). Kunming incinerators keep dioxin far below target with AI monitoring (AIBD-0691). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0691.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)

BibTeX Data Field definitions