Dryer Machine Optimization

The client is a U.S.-based company offering large-scale laundry services, operating over 11 million washer and dryer machines nationwide. They focus on automation, operational efficiency, and smart data-driven improvements to enhance customer satisfaction.

• Replaced the default controller board with a Raspberry Pi 3
• Gained full control of machine components, including:
  • Heating (4 modes)
  • Drum rotation & speed (8 levels)
  • Fan speed (8 levels)
  • Water control
• IoT-based Data Collection:
  • Sensors collect weight, temperature, and exhaust humidity during operation
  • Data sent to AWS IoT backend via MQTT protocol
• Predictive Models:
  • Developed Moisture Prediction Model with high accuracy (±0.08%)
  • Applied Fuzzy Logic to reduce drying time from 40 → 28–32 minutes
  • Applied Reinforcement Learning to further reduce drying time to 24–28 minutes
  • Achievements validated over ~160 experimental cycles

• 30% reduction in drying time using Fuzzy Logic Control
• Additional 20% reduction using Reinforcement Learning
• High-accuracy moisture prediction, with only 0.08% deviation from actual values
• Remote automation and control capability reduce on-site labor and response time
• Real-time monitoring and cloud-based analytics improve machine performance and energy efficiency