AII provides an architecture for a universal Auto-ID networked
infrastructure that can acquire, filter, model, aggregate, store,
associate, and publish massively high volumes of real-time Auto-
ID information from tagged items for use by any enterprise
application anywhere, anytime. AII provides a rich basis not
only for new tools and applications that will increase visibility
within the enterprise, but also for existing enterprise applications
that provide capabilities for detailed, real-time, item-level
analytics, control, execution, and decision making.
KEY AII REQUIREMENTS AND CAPABILITIES
Widespread Auto-ID deployments will require a new breed of
data management software and network services. As RFID and
bar-code readers interrogate billions of tags at multiple points
in the value chain, they generate vast amounts of item-level
data that must be processed and communicated across the value
chain. AII must be able to do the following:
u2022 Capture, filter, and publish data from multiple readers for individual
items and shipping units. Maintain the information
from tag reader systems that identifies individual items. Provide
context information, such as product location, shelf life,
current price, and inventory level.
u2022 Aggregate, integrate, associate, and store information about
items, pallets, cases, and shipments, including historical information.
Receive and maintain information specifying the
location and physical relationships between and among
items, including changes in those relationships. For example,
keep track of the information that a particular bottle is in a
box; the box is on a pallet; the pallet is put into a container;
and the container is put onto a ship.
u2022 Integrate seamlessly with current and future enterprise applications.
Receive and maintain smart-item information in an
extensible form, so that new kinds of information supplied
by enterprise applications u2013 such as warranty or pricing
information for items u2013 can be added and maintained without
any reprogramming of the core infrastructure.
u2022 Provide bidirectional communications and control. Publish
information to applications and devices, as well as make the
information available by query to any application or device.
Control bidirectional communications, ensuring that each
user, application, or device receives accurate and secured
information. Communicate with and directly control Auto-
ID hardware-sending, for example, updated pricing, shelf-life,
and storage information to the reader, which could, depending
on the type of Auto-ID tag, update the tag with the new
information.
u2022 Make data meaningful by processing it into useful information
and reports. Provide Auto-ID information across enterprise
boundaries in a universally accessible form u2013 such as eXtensible
Markup Language (XML) documents u2013 so that the information
can seamlessly integrate into any enterprise application
environment.
u2022 Support multiple systems. Make it easy for companies to capture
and communicate their Auto-ID data at every data
access point by allowing rapid deployment of AII across the
entire value chain, with the ability to integrate and interoperate
with existing enterprise applications and IT landscapes.
WHERE DOES AII FIT AMONG ENTERPRISE
APPLICATIONS?
AII is not a new enterprise application or database. It is a realtime
information layer that integrates with current and future
heterogeneous enterprise IT environments. AII does not make
business decisions or automate operational processes or transactions.
It is not a business intelligence application. AII captures
transitory event data from smart items across the value chain
and integrates the data into a structured model that represents
the current disposition of each smart item.
AII and SAP NetWeaveru2122
Because AII is designed to be deployed in extremely heterogeneous
environments, it can be used by multiple trading partners
with endless varieties of IT landscapes across the value
chain. AII is capable of synchronizing multiple partnersu2019 data
by using a single application across the value chain. This is
accomplished by integrating AII as an information service in
SAP NetWeaveru2122 as part of the information integration layer.
Trading partners can then access any application and back-end
system across the trading network as a Web service via the business-
object application framework layer. This SAP NetWeaver
layer makes it possible to write a single application and deploy
it in various environments, dynamically generating standard
information objects with the needed application programming
interfaces (APIs) for any trading partneru2019s IT landscape. Since
the application need only be written once instead of for each
landscape, customers can rapidly deploy and scale AII and associated
applications across the value chain, while realizing the
full value of their existing IT investments.
WHY DO COMPANIES NEED AII?
Todayu2019s enterprise systems are simply not designed to receive
and react to the huge volumes of real-time, item-level information
that can be rapidly transmitted from RFID or bar-code
readers, GPS locators, or other Auto-ID information sources.
Nor do most of todayu2019s enterprise systems track the serial numbers
of items. Directly integrating Auto-ID data into enterprise
applications can work for small, closed-loop deployments in a
single store or site. However, the enterprise application becomes
substantially overloaded as soon as the Auto-ID data needs to
be synchronized or leveraged across multiple stores, distribution
centers, warehouses, or manufacturing sites in a network.
AII is an open, generic, distributed, and independent system,
so it can handle a huge number of Auto-ID events without
burdening core enterprise software systems. The rate of information
inflow may be extremely high. For example, gate readers
at a loading dock can deliver a peak read rate of more than
20 reads per second, resulting in a peak rate of 200 reads per
second for a distribution center with 10 loading docks, not taking
into account other readers in the facility. A detailed study
of one companyu2019s requirements showed that distributing and
tracking 5 million items per day required an average rate of
378 reads per second at each location during business hours,
with readers spread across more than 1,000 distribution points.
Large volumes of information from Auto-ID readers, moving
trucks, assets, and other data sources are used to update the
AII association model, which represents the current disposition,
hierarchy, and context of all tracked items, such as individual
items, cases, pallets, containers, and shipments. Existing enterprise
systems can access or automatically receive and react to
all the information in the association model as needed and need
not deal with complex real-time events.
Individual enterprise applications generally need only a small
subset of the total information captured. The specific information
needed varies widely by application and by location in the
value chain. Rather than retrofitting each of todayu2019s enterprise
systems to capture and digest the enormous quantity of realtime
data coming from electronic sensors, it makes sense to
develop a special-purpose information layer that shields enterprise
applications from the massive quantities of information
from RFID readers, which can read Auto-ID tag data every
5 milliseconds. It also makes sense to shield enterprise applications
from the complexities of interpreting and filtering those
signals, and instead to make comprehensive, organized, and
relevant information available to those applications.
THE BENEFITS OF AII
AII provides the following operational and financial benefits.
Operational
u2022 An architecture that interprets, integrates, manages, filters,
and associates data from any source of real-time Auto-ID data
u2013 and makes that information available to any application or
device, anytime and anywhere across the value chain
u2022 The ability to manage and filter millions of bits of data u2013 and
thousands of events per second u2013 in real time in distributed
enterprise environments
u2022 Real-time tracking, association, storing, and modeling of
Auto-ID data across multiple network locations that can be
published to and updated by any integrated trading partner
u2022 Real-time, reliable, networked, and synchronized data that
enables decentralized decision making and true adaptive
supply chain planning, process management, and execution
u2022 Support for multiple applications to use, synchronize, and
update Auto-ID data across multiple processes u2013 from manufacturing
to distribution, sales, customer service, and endof-
life returns, maintenance, and support
u2022 The ability to minimize application failures or overloads by
publishing only relevant information
u2022 The information needed to employ new business models,
such as building to actual customer demand and usage rather
than building to order
u2022 Operational continuity by maintaining local Auto-ID information
for use in the event of network or communications
outages
Financial
u2022 Substantially reduced Auto-ID integration costs and time as
a result of having a single point of integration for any application,
end user, trading partner, or mobile device
u2022 Real-world knowledge that replaces assumptions u2013 enabling
managers to improve forecast accuracy and inventory availability,
to coordinate production better, and to reduce lead
times and fulfillment times for orders
u2022 Accelerated time to volume for new products as a result of
every product being completely visible throughout the entire
value chain, which makes it easier to ensure that new products
are always available
u2022 Support for mass product customization and true building
to demand by integrating customer requirements and usage
information, parts availability, and production schedules in
real time u2013 thus truly integrating customer relationship
management and supply chain management applications
u2022 Higher total return on IT investment as a result of improved
quality, speed, reliability, and availability of data within the
enterprise and across the entire value chain
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.