Internet of Things — Redefining Material Flow

Thursday, July 16, 2009

By Dr. Michael Ogle
Vice President, Educational & Technical Services
Material Handling Industry of America

To encourage the development and sharing of new ideas in the industry, MHIA’s College Industry Council on Material Handling Education (CICMHE) plans and sponsors a research focused event every two years known as the International Material Handling Research Colloquium (IMHRC). The theme of the most recent IMHRC was “The Internet of Things” and was held at the Fraunhofer Institute for Material Flow and Logistics (IML) in Dortmund, Germany. We are pleased to have IML Director, Prof. Dr. Michael ten Hompel, providing the content for this article.

The world is not predictable – isn’t it about time to design logistics systems to act like it?
A famous scientist once said that it is difficult to make predictions, particularly about the future. The supply chain world is no exception and is constantly changing in unpredictable ways. Unfortunately, we continue to build our systems assuming that we can accurately predict future events.

Complex processes like supply chains can’t be standardized in any predictable way. Only small, individual processes can be standardized. There are far too many interrelated processes and actions in material flow systems for us to assume that we can dictate where things will be and when. Flexible systems design based on the Internet of Things (IoT) takes on the unpredictability challenge and is the focus of this article.

The Internet of Data Meets the Internet of Things
When the information processing world sends your email data from one point to another, you don’t predetermine the course the data packets will take and when they will reach each server. You instead depend upon information services providers to route the data for you knowing desired starting and ending points. Control of data packets in the decentralized structured “Internet of Data” are well-known and reliably meet the world’s requirements.

The Internet of Things works in a similar fashion to provide reasonable expectations for delivery using automated and decentralized physical material flow control. You are not trying to predict specific future events. You are instead trying t to provide flexible control that guides materials through the system based on the availability of resources.
IoT provides that flexible control by using the identification capability on RFID tags along with basic service needs and priority rules saved on the tags. RFID technology supplies the key for things to act independently. It makes things intelligent, capable of communication, and lets them become smart players in service-oriented logistics environments.

The efficiency of IoT is in the interaction of the smart objects and resources. Know-how about the layout of the facility along with know-how about the desired mission (e.g. target, stopover, processing steps) of logistics objects, combined with simple data exchange and limited communication depth, allows a decentrally organized self-control of the material flow.

Learning to Live with Chance – Service Oriented Design in Logistics
IoT teaches us not to chase the dream of predictable systems, but to instead use service-oriented design in logistics that is similar to that used by computer networks.
Instead of a relatively fixed process chain, there are small atomic services which are only loosely linked with each other. Since the sequence of services is not known in advance, a large amount of flexible service chains can be represented without changing the individual services themselves.

Now the pretended certainty of a traditional supply chain management process pre-considering everything is replaced by a high flexibility design. It will be far more important in the future to guarantee the flexibility of logistics systems than to predetermine the best design for the conditions we are predicting today based on past and projected data.
Atomic services can be standardized in their functions similar to process chain elements like picking an item from a bin and placing it on a conveyor. However, the sequence of how they are deployed within a system is not predetermined and is essentially decided by the present availability and layout of physical components in a facility. This layout can be flexibly adapted to changing needs without having to change the services. Now the layout of logistics material flow systems can be changed during runtime.

Making Their Way on Their Own
Parts, pallets, and bins, as well as sets of data will in the future find their way on their own from production, to the customer, and back again for recycling. Self-reliance is the name of the game. Things will talk to each other, coordinate themselves and independently ask for necessary resources. Things have an electronic identity and will be wirelessly connected with their environment on the basis of RFID technology.

The Real-Time Challenge
There has been concern about taking information based internet models and applying them to physical flow. Can they meet the challenges of real-time systems?
Traditional control of material flow depends on real-time reaction to events (like an item moving down a conveyor that needs to be routed as it approaches a junction) within predictable times, typically within the cycle time of a conventional programmable logic controller (PLC) or similar device. The processing and reaction time of the program must be quick enough to get the item to the next desired point in the system.

Are IoT based systems up to the challenge? The same questions have been asked about the “Internet of Data”. Since the mid-nineties however, operating systems, networks and internet technologies which are not actually capable of predictable real-time response have been in effective operation. Although data packet delivery times cannot be guaranteed, events can be dealt with fast enough for the internet to function effectively. As long as resource levels are high enough and efficient rules exist for interaction between resources and data, the internet flows smoothly. The same can apply to IoT.

The Realtime Logistics project at the IML proved that the principles of service-oriented architectures as well as the RFID control of IoT can be applied to physical material flow control systems. IoT helps you embrace the reality of unpredictable material flow systems and focus on providing flexible resources and processes to take your operations to the next level of capability in an increasingly complicated supply chain.

For more information:
International Material Handling Research Colloquium
Realtime Logistics Project