Terminal Electrification & Microgrid Infrastructure Monitoring Agent with LangChain
Port terminals like Hutchison Ports Karachi are converting diesel quay equipment to fully electrified cranes and automated guided vehicles (AGVs). Managing electrified terminal infrastructure requires continuous monitoring of battery swapping cycles, charger harmonic distortion, and operational uptime across all electrified terminal assets. In this challenge, you will construct a terminal electrification monitoring agent using LangChain. The agent ingests IoT sensor feeds from e-RTGs (Electric Rubber-Tired Gantry cranes) and charging stations, flags battery degradation patterns, and auto-dispatches maintenance tickets.
What you are building
The core problem, expected build, and operating context for this challenge.
Develop a port electrification infrastructure agent using LangChain to monitor e-RTG battery metrics and automate maintenance workflows.
How work is evaluated
Evaluates predictive failure identification precision and automated work order generation.
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.
anomaly_detection_accuracy
Must flag eRTG-04 correctly.
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
false_positive_rate
Percentage of healthy cranes incorrectly flagged for maintenance • target: 0 • range: 0-100
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Build LangChain agent tools for stream ingestion of IoT sensor metrics
Implement failure pattern detection algorithms for lithium-ion terminal batteries
Automate maintenance dispatch workflow ticket generation using structured outputs
Optimize e-RTG charging schedules during peak vessel discharge windows
Reference links and supporting material
Terminal e-RTG battery telemetry dataset containing operational temperature, voltage, and charge cycle records.
Challenge setup
Evaluates predictive failure identification precision and automated work order generation.
Challenge input
JSON array of crane telemetry records (battery temp, cycle count, cell delta V)
LangChain
Required framework for building tool-calling agent executors.
Evaluated output
JSON work order with target crane ID, anomaly type, and severity score
- Must flag eRTG-04 correctly.
- Percentage of healthy cranes incorrectly flagged for maintenance • target: 0 • range: 0-100
- Benchmark: IoT-Predictive-Bench
- False Positive Rate target: 0%
- 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
- LangChain · Required
- Langchain · Optional
- aiXplain · Optional
- LangChain · Required
- Langchain · Optional
Evaluation contract
- anomaly_detection_accuracy · Weight 1
- false_positive_rate · Weight 1
Metrics
- false_positive_rate · Target 0
Benchmarks
- IoT-Predictive-Bench
- general
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON array of crane telemetry records (battery temp, cycle count, cell delta V)
Output: JSON work order with target crane ID, anomaly type, and severity score
Execution harness: Python
Evaluation tools
- LangChain · Agent Framework
Recipe freeze marker
No recipe freeze time is recorded.