 | Level: Introductory Luc Chamberland (lchambers@ca.ibm.com), Program Manager, IBM Mike Starkey (starkey@ca.ibm.com), Senior Technical Staff Member, IBM
01 Aug 2003 In our series about integrating business processes to streamline the supply chain, this first article gives an overview of supply chains, and describes how suppliers and customers can use global repositories to synchronize data. The authors explain supplier and customer architecture, and describe item synchronization between suppliers and customers. The article also outlines how to use products in the WebSphere family to take either a buy-to-integrate or build-to-integrate approach to the scenario.
Introduction
In many industries, such as retail, there's nothing new
about needing an effective supply chain. Evolving the integration and automation of
business processes, with your trading partners and within the company,
is essential to increasing time to market and lowering overall
costs. Whether a customer works with a
few dozen suppliers or perhaps thousands, the ability to manage and automate
the flow of item data between them quickly and reliably could spell the
difference between success and failure.
Moving from paper-based to electronic systems to
handle purchase orders, bills of lading, invoices, and the exchange of item
information continues to define a company's competitive edge.
EDI-like systems have been available for over
30 years, and Internet-based electronic exchange is now growing quickly.
This article discusses how suppliers and
customers can use a global repository to manage the synchronization of item
data. For our purposes, a customer is a business unit that purchases goods or services, such as a retail chain
purchasing from their suppliers. A supplier is a business unit that
supplies goods or services, such as a consumer packaged goods (CPG) company. The data is stored
privately within supplier and customer datastores,
but is shared using a global repository to store the item data in a common
agreed-to interchange format. Because automating this process typically also
requires some intra-enterprise application integration, we will also examine
how to effectively automate internal processes.
Item synchronization between suppliers and customers
Imagine a scenario where a large customer is dealing with
thousand of suppliers. The customer
wants to consolidate all their item data into its own systems to be able to
act, in turn, as a supplier in the next stage of the supply chain. This requires
that the customer receives item data from each of its suppliers.
Item synchronization is the
process of acquiring item data and keeping it up-to-date between suppliers
and customers. All suppliers have their own internal systems where they keep
the item data. This information might be
provided today on paper, on the Web, through EDI,
or some other electronic means. The customer typically has its own systems where it needs to store the item information, so reconciling all these approaches
could essentially mean that the customer has to re-key much of the supplier's
item information into its own systems. The time and cost to re-enter item information could be
significant, impacting the dustomer's ability to introduce items to market
quickly. And, if the customer is manually re-entering item information,
errors could be introduced. The wrong
SKU could be ordered, which could cause stock problems to arise, which in turn
could cause advertising campaigns to be out of sync with item availability -- a
nasty chain of events, but one which is unfortunately all too common.
An approach to help the customer limit such pain is to get
supplier item information from a global repository. The global repository serves as a centralized
repository for standardized item information, as shown in Figure 1 below. All item information published to
the global repository follows the same format. The consistency of this format,
and the method of accessing global repository data, mean that the customer can
much more easily automate the loading of supplier data into the customer's
catalog. Once a customer chooses to use a
global repository for item synchronization, suppliers that also choose to
publish information to that global repository may enjoy a competitive advantage
over those that don't.
Figure 1. Global repository to synchronize data
UCCnet, a subsidiary of the
Uniform Code Council, is an example of a global repository that has built a
trading community to better enable item synchronization across supply
chains. UCCnet
supports Internet protocols such as HTTP and AS2, and the UCCnet
data pool enforces item standards to help ensure data integrity in the supply
chain. The item content refers to item-related data (for example, description), metadata, and other attributes and characteristics. Using XML schema, the UCCnet item format has several dozen standard item descriptors. Without such
industry-standard formats, the time required by customers to manage
data coming from multiple suppliers would be significant, in addition to the
number of errors introduced and downstream implication of such errors (stock-outs, accounting errors, and so on).
As other global repositories also
develop generic formats, they can share these formats amongst themselves so
that supplier items could be registered to an even wider audience. This is the promise of global repositories:
as aggregators of product information, they accelerate the supply chain and
increase market awareness.
Supplier architecture overview
To work with a global repository, suppliers need to
implement solutions that produce and publish global repository-compliant item
descriptions to the global repository, while customers retrieve and consume
relevant descriptions from the global repository. Because suppliers and customers only interact with the global
repository and not with each other, in the case of item
synchronization the concerns of each participant can be
separated. Suppliers and customers still interact to agree on pricing, packaging,
and possibly item selection.
Figure 2 below shows a high-level
solution architecture from a supplier's perspective.
Two key portions of the architecture are shown:
- The B2B portion, where item synchronization with the global repository takes place.
- The B2C portion, where the supplier's item catalog is made available through a portal.
The back-end systems, or enterprise information systems (EIS1,
EIS2, and EIS3), represent different parts of the supplier's business -- EIS1 for
item information from product development, EIS2 for pricing information from
the business unit, and EIS3 for availability information from the manufacturing
unit. The supplier would like to
protect its investment in these mature systems (such as CICS, SAP) by using
business processes to manage and consolidate its data.
Figure 2. Architecture overview for a supplier
In the B2B portion of the supplier solution, the EIS1
back-end system (perhaps a CICS or SAP system) holds item descriptions developed
by the product development group. When an item is ready to be published, an
asynchronous message is sent through an adapter to a business process
engine. This transformation engine
launches the data cleansing process that converts the description into the
format required by the global repository. Proprietary terms and jargon, such as intra-enterprise code names, are
removed and general information required by the global repository standard is
inserted. If exceptions are caught that
cannot be easily corrected by the cleansing flow, then a workflow activity is
launched. A data entry supervisor makes
the required fix and approves the item description for publication. In most cases, the item will require approval
before it is published to ensure the integrity of the data. Once an item
description has been cleansed, it is published out through a gateway to the global
repository.
The supplier will likely want to have a non-automated path
for customers to find out about new items. Also, because the global
repository has a standardized format, the supplier may want to provide
additional information, such as pricing or availability, about the items in some
form. This data comes from other internal systems (for example, EIS2 and EIS3) and
therefore an aggregation of item data is required and a complete description of
the item created.
To create the supplier's catalog, information from the three
separate back-end systems needs to be aggregated. The transformation engine retrieves these
pieces using adapters, aggregates them, then primes the catalog database.
The aggregation of the data should include the cleansed item data that was
stored in the global repository so that there is a common set of information
between the global repository and the supplier's catalog. Because the
identifier of the item is likely different in the global repository than in the
supplier's system, and because the identifiers may be different among the supplier's
systems, creating the aggregated item information is further
complicated by the need to map the identifier in one system to that in another
to get the related information. Integrated portlets can be provided to customers with
a personalized view of the supplier's catalog when they browse.
Suppliers could benefit significantly from monitoring the
overall effectiveness of their solutions (for example, by understanding the latency of the
data cleansing and publishing process). To help a supplier monitor a variety of business-level metrics, instrumented probes can be inserted in specific runtime artifacts where event data needs to be captured. The event data can then be
processed to render business metrics in a dashboard.
Customer architecture overview
The customer architecture is much like a mirror image of the
supplier architecture. The customer can get a list of item descriptions from
the global repository. The items of interest are selected and then need to be
stored in the customer's systems. However, in most cases the item data needs
to be massaged and usually augmented before storing with customer-specific
information. Some data transformation can be automated, although usually
someone needs to augment and approve the item data before it is stored, to
ensure the integrity and completeness of the internal data.
An IBM WebSphere product implementation
There are a variety of ways to implement the scenario we've outlined. The WebSphere family of products lets you take either a buy-to-integrate or
build-to-integrate approach.
WebSphere Business Integration
If the target global repository already exists and you simply need to connect to it, the WebSphere Business Integration products provide a robust solution with customizable templates and
connectors that can have you up and running out of the box very quickly. WebSphere includes
industry-specific solutions that have been tailored for
suppliers, customers, global repositories and the business problems faced by
participants in that industry.
With WebSphere Business
Integration, you can avoid the transactional costs associated with value-added network (VANs) by using EDI-INT AS2 to
communicate with your trading partners. Some retail customers or global repositories are requiring their
suppliers to implement AS2-based solutions.
The first step in item synchronization is for the supplier
to quickly leverage item data from its existing back-end systems (EIS1 in Figure
2). WebSphere
Business Integration offers a variety of back-end adapters, including adapters
for i2, PeopleSoft, Oracle, and other systems.
To manage the transformation of item data into a format that conforms to
that of the global repository, WebSphere Business
Integration provides a robust transformation engine that can handle automated
straight-through processing and flows requiring human interaction.
Process templates are provided to accelerate
the flow development. The bundled MQWorkflow product
lets you define complex human interaction flows, required in our scenario to
handle item processing exceptions.
The IBM Item Synchronization for Suppliers solution builds
on top of WebSphere Business Integration and offers
function specifically geared to enabling connectivity to
UCCnet. You can share data with your partners, and also validate,
reformat, and automatically send item data to the UCCnet
registry when changes to item data are made in your ERP.
The UCCnet-certified
product includes reusable process templates that facilitate item synchronization.
One of the largest supermarket chains in the U.S. addressed their item synchronization needs
with WebSphere Business Integration.
They've begun to see the results of reduced
price discrepancies, improved workflow, and fleet optimization, which contribute
to cost and time savings, better time-to-market, and more flexible processes.
WebSphere Application Server and WebSphere Studio tools
To build a custom solution that is solidly based on key
Internet standards, IBM WebSphere Application Server
and the tightly integrated WebSphere Studio tools, as seen in Figure 3, offer a robust and flexible build-to-integrate solution.
Figure 3. WebSphere Studio tools
WebSphere Application Server
delivers several critical features to help your solution integrate with a global
repository. WebSphere Application Server provides a services orientation to
facilitate a loose coupling of services across enterprises, strongly supporting
J2EE and Web services standards. Your
time-to-market is reduced by the support for code reuse, and there's minimal disruption
to existing applications when changes to individual services are required.
WebSphere Application Server
handles the transformation and flow management requirements of the solution by:
- Letting you dynamically manage interactions between business processes with the
integrated Web services workflow engine, and
- Providing users with the
flexibility to dynamically interact with your solution by building and
submitting runtime queries.
WebSphere Application Server
Network Deployment includes a Web services gateway, which provides a secure
environment to transmit data through your enterprise firewall.
Use the gateway to transmit to and from the global
repository, handle protocol conversions (for example, HTTP to JMS), and manage authentication
and authorization of users.
To build your integrated B2B application, the
WebSphere Studio Application Developer Integration Edition
tools, built on the open source Eclipse framework, let you:
- Create adapters to access back-end systems
- Use WebSphere Business Choreographer to create business
processes to link a variety of activity-based services
- Integrate human workflows
- Build J2EE artifacts
- Unit test and debug an end-to-end solution.
Personalized view of the supplier catalog
To effectively browse the supplier's catalog, customers
want:
- All relevant product information, such as product description,
price, and availability.
- The information filtered by the
nature of their account, such as platinum or gold, and buying preferences.
First, you need to pull together all the relevant information
from the back-end systems. Use the same
adapter technology described above to access those systems, then use the
appropriate flow technology (WebSphere Business
Integration or WebSphere Studio Application
Developer, Integration Edition) to aggregate the data into a single file.
Next, use WebSphere
Commerce Business Edition to build and deploy your product catalog, enabling it
for browsing, personalization, shopping, and payment management.
WebSphere Commerce Enhancement Pack includes
commerce-enabled portlets that let you harness the
personalization features of WebSphere Portal Server
to your catalog. After logging onto the supplier's
portal, the customer immediately gets information customized to their
preferences.
Conclusion
Global repositories help suppliers and customers automate
the supply chain, making it more reliable, efficient, and cost-effective.
The WebSphere family of products provides the capability for suppliers and customers to
effectively build, deploy, and manage robust and scalable solutions.
Item synchronization is just the first step.
Other aspects of collaborative commerce
include:
- Inventory and price synchronization
-
Optimize inventory turns by sharing pricing and availability
information.
- RFQs and reverse auctions
-
Enable customers to request quotations from suppliers and even participate
in a competitive bid.
- Promotion
synchronization for shared marketing campaigns
-
Enhance the exchange of information between suppliers and customers for
special promotions.
- Scan-based trading
-
Use the customer's daily POS scan data to automate payments for purchase
order fulfillment and procurement.
- Business intelligence
-
Leverage event data from trading partners to trigger and define a variety
of activities, including: inventory management; fulfillment optimization;
business-level monitoring of both intra- and extra-enterprise; and handling
unanticipated business events. All willl enable your business to sense and respond in a proactive way.
Resources
About the authors  | 
|  | Luc Chamberland is Program Manager of WebSphere Business Scenarios. He works with various IBM teams to understand how customers use WebSphere products together, and provides recommendations to product groups on platform integration. Previously, Luc managed teams in the IBM Toronto Lab working on XML parsers within the Open Source community, and on the IBM VisualAge for Java product. He has written and presented on Java tools, XML, and Open Source. You can contact Luc at lchambers@ca.ibm.com. |
 | |  | Mike Starkey is the lead architect in the WebSphere Business Scenarios Development team. He has focused his career on understanding how customers build complex, distributed enterprise solutions, and designing
products to meet their needs. |
Rate this page
|  |