Intermodal shipping today has thousands of players including the operators of ships, trains, trucks, and airplanes.
The inclusion of a large number of ports and diverse shipping depots results in a complex logistics system is far from as efficient as it might be. By using remote sensing and big-data analysis, the Internet of Things (IoT) has the potential to make shipping logistics more seamless, transparent, and efficient.
Logistical efficiency is closely linked to visibility — knowing where a shipment is and where it’s headed at any given time. The critical technologies for improving in-transit visibility are cloud-based GPS to report location and Radio Frequency Identification (RFID) technologies that identity and locate shipments down to the pallet level.
Data gathered from GPS and RFID technologies will do more than enable shippers to predict the time of arrival exactly. Deploying other types of sensors will enable big data technologies to sort and parse the data from sensors monitoring the environment within the transport vehicle. Shippers can monitor parameters such as temperature, humidity, and vibration that may impact the quality of a product in-transit.
Data can be transferred into the cloud using a small data aggregation system and communications system in the train car or truck. Every stakeholder in the supply chain would be able to identify the pallet and know its position using GPS coordinates, weather, and traffic conditions. Vehicle-oriented information such as a truck driver’s driving pattern or a ship’s average speed could also be reported.
For high-value shipments subject to hijacking such as pharmaceuticals, precious metals, and state-of-the art microprocessors, digital cameras might be employed. An agile supply chain will also empower shippers to reroute shipments at will and create an on-demand maintenance program that responds to data from sensors embedded in engines and other critical components in the vehicle.
Keys to the success of the system are the cost of the components and the robustness and longevity of the on-bard system. If millions of RFID transponders and sensors are to be used, they must be inexpensive. In addition, the data aggregation and transmission systems must be low cost and energy efficient.
Global data flow
The first hop in transmitting shipment tracking data is from an RFID tag on the pallet (or other source such as a temperature sensor) to a data aggregation system. Advanced systems employ active RFID tags that periodically communicate 'I'm here' status information (typically once every 10 seconds) to an RFID reader located in the truck, train, ship or warehouse where the pallet happens to be. Active RFID tags operate in two bands to attain the appropriate range. The 2.4-GHz band has a range of 50 meters while the 868 MHz band has a 500 meter range.
The reader has two other communication partners: GPS units that tag GPS position to the inventory data; and typically a GPRS (General Packet Radio Service) radio that relay the information to cloud servers.
GPRS is a packet-oriented mobile data service that utilizes the 2G and 3G cellular networks. Usage is typically charged based on volume of data transferred — as opposed to circuit-switched data, which is usually billed per minute of connection time.
The use scenario of hundreds of RFID tags continuously communicating with the reader obviously raises concerns about data traffic congestion and energy consumption. The design team can address both concerns with an event management engine and an ultra-low power radio protocol.
The design goal is for the data reader/aggregator to accommodate several hundred tags in close proximity. The reader/aggregator typically runs off the truck battery or power supply on the train or ship. To keep operating cost low, the tags must run CR2032 coin cells (watch batteries) for not less than five years. The key to the battery life requirement is in controlling the application duty cycle.
The RFID chips should be able to wake from — and drop back into — an ultra-low power sleep mode with a few milliseconds. Energy consumption in sleep mode should be about 900nA).
Low-power sensor networks
In addition to location tracking data, an advanced intermodal shipping supply chain requires data from a low-power sensor network. The most significant engineering challenge is the seamless integration of the “RFID network” and the sensor network.
Temperature readings and similar environmental data are well-suited for collection by wireless networks. On the other hand, data for automated maintenance programs (to monitor the health of a truck engine, for example) may not be suitable a wireless network.
Wherever possible, wireless connectivity is desirable – and from that perspective, the same design goals apply as for many sensor networks. Although several approaches to a robust solution are available, a typical architecture for ultra-low-power sensor networks is based on a mid-range MCU such as one from Texas Instrument’s MSP430 family and a multi-protocol 2.4 GHz transceiver such as the TI’s SimpleLink CC2500.
In this solution, the MCU runs both the RF protocol and application layers, which allows the application to have more direct access and visibility into the RF and physical layers. Pushing the intelligence into the MCU allows for a simple and robust radio doing what it does best: transmit and receive wireless data.
Since the MCU usually has the most resources in terms of memory, processing power, as well as digital and analog integration, this configuration is also well adapted to many types of wireless protocols and applications.
Conclusion
Integrating technologies developed for IoT-like wireless sensor networks into moving cargo vehicles will have far-reaching effects on intermodal shipping. Among other benefits, it will create fleet efficiencies, save fuel costs, ensure in-transit temperature stability and reduce asset loss in terms of both cargo and rolling stock. RFID and GPS technologies that are already in wide use will have to be efficiently integrated with data from other sensors. There are numerous architectural approaches to this challenge. Successful implementations will include both ultra-low power operation and a tiered communications network robust enough to operate in challenging environments.