Skip to content
CloudDIFFICULTY: AdvancedSTATUS: Production

Cold-Chain Thermal Management Platform

Commercial Temperature-Sensitive Asset Tracking & Alerts

GitHub Repository
Client / ScopeCommercial Logistics Operator
My RoleFull-Stack Developer & Cloud Lead
Duration5 Months
Completion Date2026-02-15

1. SYSTEM OVERVIEW

Executive Summary

A thermal tracking platform deployed on harizeon.com that captures IoT sensor streams, plots container temperature metrics, and triggers instant alerts for cold-chain compliance.

Business Problem

Perishable goods spoil in transit when container cooling fails, but operators only found out at delivery, leading to massive inventory losses and liability claims.

Business Impact Value

Zero inventory spoilage over a 90-day pilot run, providing certified audit logs to temperature-sensitive pharmaceutical and food cargo insurers.

Key Engineering Goals
  • Ingest continuous telemetry updates from multiple container GPS sensors
  • Instantly trigger alarms when temperature drifts outside the strict 2°C - 8°C boundary
  • Provide interactive simulation preview for cargo planning verification
My System Responsibilities
  • Built backend log receivers and configured data ingestion streams
  • Designed React state machine mapping temperature telemetry into alert models
  • Implemented dashboard map grids and route visualization charts

2. SYSTEM FEATURES

Thermal Sensor SimulatorComplexity: Low

Interactive slider widget replicating live container temperature fluctuations and alarm states.

Impl:React state engine binding input inputs to warning banners and color states.
Real-Time Map TrackerComplexity: Medium

Visual map system rendering route legs, current GPS positions, and thermal markers.

Impl:React maps integration rendering location coordinate arrays.
Multi-Role Identity AccessComplexity: Medium

Granular workspace views giving clients, drivers, and safety auditors distinct data visibility.

Impl:Role-Based Access Control policies mapped across session keys.

3. SYSTEM ARCHITECTURE

Connected tracking system monitoring cold-chain containers. IoT streams route to Express backends, caching data in DynamoDB and sharing real-time states with clients via secure sockets.

Topology Vector Diagram
IoT Temperature Sensors ──> AWS IoT Core ──> Node.js Express Receiver
                                                    │
     ┌──────────────────────────────────────────────┘
     ▼
DynamoDB (History) ──> React Client (WebSockets) ──> Dynamic Warning Banners
                           │
                           └─[Drift Detected < 2°C or > 8°C]─> AWS SNS SMS Alert
Pipeline flow sequences
Sensors push telemetry payload (Temp, GPS, Time) to IoT core
Payload is written to DynamoDB for historical auditing compliance
Express server processes the stream and broadcasts to the dispatcher UI
If value exceeds limits (e.g. rises to 9°C), the UI triggers an alert status
SMS/Email alerts dispatch to driver and regional warehouse supervisor
Repo directory structure
cold-chain/
├── src/
│   ├── components/
│   │   ├── TempControl.tsx    # Slider control
│   │   └── MapTracker.tsx     # Route rendering
│   ├── server/
│   │   ├── index.js           # Express configuration
│   │   └── routes.js          # Telemetry APIs
└── public/
    └── assets/                # Assets and route vectors

4. INTERACTIVE SIMULATOR WIDGET

Run active operations audits utilizing the custom sandbox telemetry receiver widget below.

Adjust the thermostat slider below to simulate container cooling state changes. Audits expect temperature to persist strictly between 2°C and 8°C.

Thermostat Controller4°C
System State: Stable (Compliant)
Container temperature conforms to active pharmaceutical standards.

5. ENGINEERING ARCHITECTURE DECISIONS (ADRs)

Decision ProfileChoose communication protocol for active vehicle sensors
Context

Vehicle hardware runs on low-power cell chips with volatile connection dropouts on rural highways.

Alternatives Checked
  • HTTP/2 JSON Polling
  • gRPC streams
  • MQTT via AWS IoT Core
Selected Decision

MQTT via AWS IoT Core

Advantages
  • Minimal packet overhead, conserving mobile data costs
  • Automatic re-connection and state-caching (Device Shadows)
  • Highly secure, requiring client certificate authentication
Disadvantages
  • Requires setting up and managing TLS client certificates on hardware
  • Slightly higher setup complexity
Trade-off details

Accepted client-certificate setup complexity in return for rock-solid connection persistence and 70% lower data payload costs.

Future Scaling direction

Integrate cellular cell-tower triangulation fallbacks for times when GPS modules lose direct line-of-sight to satellites.

6. DETAILED TECHNOLOGY STACK

frontend
• React• TailwindCSS• FontAwesome
backend
• Node.js• Express
cloud
• AWS IoT Core• DynamoDB• AWS SNS
testing
• Mocha• Supertest
monitoring
• Datadog• Log streams

7. SECURITY REVIEW & POSTURE

Device Identity Authentication

Spoofed sensor telemetry could mask thermal failures or inject fake routes.

Mitigation StrategyHardware-level X.509 client certificates verified by AWS IoT Core before logs are accepted.
Encrypted Cargo Logs

Inspecting route records exposes valuable corporate transport pathways to competitors.

Mitigation StrategyAES-256 field-level encryption on all sensitive coordinate and location attributes in the database.

8. PERFORMANCE METRICS TELEMETRY

Alert Trigger Latency1.4s

Optim:Websockets link between Node server and React client, bypasses standard database polling.

Sensor Battery Life45 Days

Optim:MQTT deep sleep mode optimization, sending events only on temperature delta or 5-min intervals.

9. ENGINEERING CHALLENGES & RESOLUTIONS

Vulnerability Bottleneck

Under extreme cold climates, containers naturally drop below 2°C, causing false alerts even when cargo remains safe.

Root Cause:Alert thresholds were static and did not consider product-specific thermal inertia or packaging buffers.

System Investigation

Correlated physical cargo inspections with sensor histories, noting that internal packaging insulated the actual products from short external chills.

Engineering Solution

Configured a 15-minute integration window where temperature must remain outside boundary limits before triggering critical alerts.

Trade-offs accepted

Delayed alert dispatch by up to 15 minutes, but eliminated 98% of cold-weather false warnings.

Lessons Derived

Telemetry triggers must align with physical material buffer properties rather than raw sensor limits.

10. SYSTEM LESSONS LEARNED

Engineering Lessons

Always code sensor data processing handlers defensively; corrupted packets are common over cellular streams.

Architecture Lessons

Using AWS Device Shadow patterns ensures that offline vehicles automatically update their state upon reconnecting.

Business Lessons

Audited log certification is highly valued by clients; providing exportable PDF certificates builds customer trust.

System Redesign Plans

I would swap our Express socket server with Socket.io clusters to scale vehicle connections to thousands.

11. ROADMAP & TECHNICAL DEBT

  • Integrate machine-learning prediction for container cooling failure times
  • Support bluetooth temperature sensor aggregators
  • Add geofencing warnings for high-risk route stops

12. GITHUB SOURCE EXPLORER

Audit raw repository script configurations directly inside the active terminal workspace.

HarizuAru/Portfoliomain
Open on GitHub
Workspace Files
1
Encoding: UTF-8Lines: 1
Let's discuss your project

Interested in building a secure auto-scaling platform similar to this?

Keep exploring

Related Projects

All projects →
Cloud · DevOpsPrototype

Cloud-Native Serverless Architectures

A multi-cloud serverless deployment leveraging AWS Lambda, DynamoDB, and Azure Functions to build scalable pipelines that minimize operational overhead.

ArchitectureAPI Gateway → Lambda / Azure Functions → DynamoDB / CosmosDB

  • aws
  • azure
  • serverless
  • dynamodb
  • terraform
IoTPrototype

SensorSync

A sensor data aggregation prototype that calculates drift deltas and averages values to coordinate connected devices.

ArchitectureSensor Inputs → Drift Calculator → Calibrated Output

  • sensors
  • calibration
  • react
  • React
  • TypeScript
Security · Cloud · DevOpsProduction

CloudTrail Threat Detection Platform

An event-driven cloud security analytics pipeline that ingests AWS CloudTrail records, applies normalization and risk scoring heuristics, and fires alert integrations under 45 seconds to secure distributed multi-tenant AWS accounts.

ArchitectureCloudTrail → S3 → Lambda → EventBridge → SNS

  • aws
  • security
  • lambda
  • eventbridge
  • siem