Build Construction Delay & Procurement Overrun Predictor with AutoGen
Major urban adaptive reuse projects like King's Cross face potential schedule delays and material cost overruns due to supply chain volatility and contractor bottlenecks. Developers will construct a multi-agent construction management predictor using AutoGen. The system pairs a Supply Chain Monitoring Agent, a Weather Analysis Agent, and a Schedule Risk Coordinator Agent. The agents analyze site logs, procurement purchase orders, and weather forecasts to predict project completion delay risks and calculate contingency budget reserve adjustments.
What you are building
The core problem, expected build, and operating context for this challenge.
Develop an autonomous construction schedule delay risk predictor using AutoGen multi-agent collaboration.
How work is evaluated
Evaluates predicted delay days and recommended contingency reserve adjustments.
Challenge datasets
Public datasets, plus the private datasets you have access to.
How submissions are scored
These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.
Contingency Math Precision
Ensures required contingency strictly equals delay_days * daily_overhead_cost
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
Delay Prediction Accuracy
Precision of total predicted delay days calculation • target: 0.95 • range: 0-1
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Configure AutoGen agents for multi-factor construction schedule analysis
Build custom delay prediction tools combining weather impact and material delivery logs
Calculate recommended contingency financial reserves based on predicted delay days
Generate automated progress reports for project owners and general contractors
Reference links and supporting material
Dataset of 40 construction project site logs, material procurement timelines, and weather forecasts.
Challenge setup
Evaluates predicted delay days and recommended contingency reserve adjustments.
Challenge input
JSON with steel_delivery_delay_days, adverse_weather_forecast_days, baseline_completion_days, daily_overhead_cost
AutoGen
Required framework for multi-agent risk prediction.
Evaluated output
JSON with total_predicted_delay_days, critical_path_impacted (boolean), required_additional_contingency
- Ensures required contingency strictly equals delay_days * daily_overhead_cost
- Precision of total predicted delay days calculation • target: 0.95 • range: 0-1
- Benchmark: ConstructionDelayBench
- Delay Prediction Accuracy target: 0.95
- 1 public reference case
- Python execution harness
- Python sandbox (unavailable on Versalist)
Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.
View technical recipe
Configured tools
- AutoGen · Required
- Voiceflow · Optional
- BoTorch · Optional
Evaluation contract
- Contingency Math Precision · Weight 1
- Delay Prediction Accuracy · Weight 1
Metrics
- Delay Prediction Accuracy · Target 0.95
Benchmarks
- ConstructionDelayBench
- general
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON with steel_delivery_delay_days, adverse_weather_forecast_days, baseline_completion_days, daily_overhead_cost
Output: JSON with total_predicted_delay_days, critical_path_impacted (boolean), required_additional_contingency
Execution harness: Python
Evaluation tools
- AutoGen · Agent Framework
Recipe freeze marker
No recipe freeze time is recorded.