The object related storage approach has the
following advantages:
Real-time data collection. Because data storage
occurs object related, events that take place in the real
world can be collected right where the action takes
place even if network infrastructure is unavailable. For
example, in a cold chain, data on environmental factors
such as temperature and humidity can be collected
close to the object via sensors. Even if the cold chain
has to be kept intact across several logistic stages, the
data that is necessary to verify the cold chain is exactly
where the object is located. In each logistic stage, we
can determine whether the cold chain has been violated
because environmental factors increased or decreased
across a certain time period (“the oysters are uneatable
because during the past 5 hours, the temperature in the
container has risen above 10 degrees Celsius for at
least three times for 5 minutes”). Without the data-ontag
storage, we would need infrastructure for each and
every logistic stage to monitor the environmental
factors. At the end of the cold chain, all this data
would have to be combined in order to detect
irregularities. It would take time to determine whether
the cold chain has been interrupted and even more time
to initiate counteractive measures such as repeating an
order. In most cases active tags are needed for realtime
data collection (Class 4 transponders).
Support of unmanaged processes. Processes that are
not centrally controlled can be supported by object
related data. For example, routing information that has
been stored close to the object can be used to
decentralize internal logistic processes. Similar to lead
codes on envelopes or parcels, objects with routing
Proceedings of the 40th Hawaii International Conference on System Sciences - 2007
4
Proceedings of the 40th Annual Hawaii International Conference on System Sciences (HICSS'07)
0-7695-2755-8/07 $20.00 © 2007
information can be automatically controlled across all
levels of production and storage. Additional handling
information can improve the quality of the unmanaged
logistic processes by taking into consideration
specifics of transported goods such as temperature or
fragility.
Relief of central systems. Particularly when large
amounts of data need to be processed simultaneously,
the decentralized pre-processing of data can help
relieve the central system [21]. For example, the
ripening of particular goods could be monitored by
sensors which collect data such as ph and sugar levels
and environmental factors such as temperature and
humidity. With the help of simple logistics,
information concerning ripening can be aggregated and
the central control of such processes can be relieved.
Increase of stability. In a centrally controlled
system, the failure of the central computer results in
the failure of the entire system. If we decentralize the
object related controlling processes (as described
above), we can increase the stability of the entire
system because the failure of one single system does
not result in an overall failure.
In the following, we briefly introduce the
disadvantages of the data-on-tag concept:
Object dependant data availability. Data storage on
tags attached to the object implies that this data is only
available if the corresponding object is within the
reach of the respective infrastructure.
Storage capacity limitations. Although first
prototype RFID tags which can store several megabits
of data have already been developed [22], at this time,
however, the storage capacity is generally too low to
collect all the object relevant data on a tag. Therefore,
some of the data has still to be stored in central
databases.
High Costs. In order to store data with the object,
expensive tags with high storage capacities are
necessary.
By storing the object relevant data on the tag, new
ways to protect and secure data need to be taken into
consideration [23]. Because the collected data is with
the object and can only be accessed if one has the
object at hand, the owner of the object is
simultaneously also the owner of the collected data. In
general, for object related data that includes
manufacturing date, expiration date, information on
how to handle the object, the situation is
unproblematic and even desirable. Data that has been
collected during the manufacturing process of the
object and that give insights into business know-how
are more problematic. If the current owner of the
object is not the manufacturer of the object, this data
should not be available. If an object changes owners