Approaching Manufacturing, Retail, and Distribution with Supply Chain Integration

The difference between supply chain integration and application integration is in their relationship to one another. One is the rail, the other is the train. Application integration, as we've discussed, requires placing new approaches and technologies around the process of extending the reach of applications, enabling them to exchange information with other applications that exist in other organizations. Supply chain integration represents the enabling processes that run on top of the infrastructure that application integration creates (see Figure 17.2). Whereas application integration is about a tactical process that depends heavily upon technology, supply chain integration is more about strategy.

Figure 17.2. Application integration creates the infrastructure that allows supply chain integration (SCI) to work.

graphics/17fig02.gif

The symbiosis of supply chain integration and application integration is clear. The technology, the approaches, and even the benefits are much the same. A tightly coupled supply chain is dependent upon the success of application integration, along with such application integration technology as message brokers, application servers, and standards such as XML, EDI, ebXML, UCCnet, and RosettaNet (covered in Part III of this book).

Supply chain integration is a well-known and time-proven concept. There is significant literature dealing with it, so I will not cover it exhaustively here. My purpose is to understand the basic notion of supply chain integration and its usefulness in the context of application integration.

Value of the Chain

Supply chain integration may well represent the Promised Land. But as with most Promised Lands, the road leading to it is hardly smooth or straight. A substantial commitment to technological and organizational change must be made for supply chain integration to work. Companies with little corporate patience, companies with little corporate vision, and companies looking for short-term gains need not apply. There is risk aplenty on the road to this Promised Land. This glory bus is only for organizations willing to confront the risk, willing to face new and changing technological approaches, willing to absorb significant short-term expenses, and willing to invite the need to layer into other organizations not under their direct control other organizations that could even include a competitor or two.

No foolproof method for integrating a supply chain exists. However, organizations can take steps to ensure that they minimize the risks. The smart strategy is to plan for the long term, implement in the short term, and "keep your eye on the ball" the business reasons for supply chain integration. The most difficult trick may be keeping the technology in perspective. It is there to be controlled, not to be worshipped. Ultimately, it is the tool, not the master. Finally, organizations must make wise decisions. They must avoid the carnival barkers in the technology marketplace. They need to base their decision making on what adds the most value to the chain, not on what glitters or what is trendy. You'll find the term "Supply Chain Acceleration" attached to many products these days; few have anything to do with supply chain integration.

It is a fool who does not respect the risks involved. Every organization should feel the weight of these risks. However, the remarkable opportunity that supply chain integration represents is beginning to outweigh the risks. We are fast approaching a point of critical mass where the greater risk might well be in failing to integrate.

Organizations can no longer afford the luxury of perceiving their operations as if they existed in a vacuum. No organization is an island, to paraphrase a quote. Every organization needs to collect comprehensive, accurate, and timely information throughout the supply chain.

Once this information is gathered, it must be analyzed in order to better comprehend the causes and effects of the business environment on the core business. Once such an analysis has been accomplished, the resulting knowledge will allow the organization to make informed business decisions and to utilize information as a mechanism to gain market share.

Supply Chain Entities

Within the notion of a supply chain there are several entities, including:

  • Supplier

  • Consumer

  • Supplier systems

  • Consumer systems

  • Private information

  • Public information

  • Private processes

  • Public processes

Suppliers are those organizations that supply a good or service as part of the supply chain. An example would be lumber providers for housing construction. It's the role of the supplier, in a supply chain, to accept an order and provide information back as to how the order will be completed. Suppliers typically receive payment.

Consumers are those organizations that use the good or service to meet some business need, such as building a car or perhaps providing logistics services. It's the role of the consumer to place orders, providing order information. Consumers typically pay suppliers.

Supplier and consumer systems are those systems that consume and produce information to carry out the defined roles of the supplier and consumer. These can be mainframes, Enterprise Resource Planning (ERP) systems, client/server systems, Web-based systems, or anything that can produce information of value to the supply chain and receive information as appropriate.

In addition to the information systems themselves, you also have the notion of private and public information. Private information is information that is not to be shared within the supply chain; examples would be employee information or sales data. It is critical when dealing with a supply chain that you mark data as private or else risk having sensitive information get outside of the organization. In contrast, public information is information that is sharable and not sensitive, such as order information or payment verifications.

In addition to information, you also have the concepts of public versus private processes. Like public and private information, private processes are processes that are intracompany and do not have anything to do with the supply chain. Public processes are shared, with participation by multiple organizations. Public versus private processes are very important in the world of application integration because they map to Process-Oriented Application Integration and are controllable within the process engines found in prevailing application integration technology. What's more, they may link to emerging and existing standards that define processes for supply chains, including RosettaNet and ebXML.

Defining Your Supply Chain

Supply chains support the flow of goods and services from their origin to their endpoint the customer. The components of the supply chain may include the original suppliers, multiple production operations, logistics operations, retailers, the customer, and even the customer's customer. For example, an organization that builds birdhouses has a supply chain that includes the lumber company that processes trees into lumber, the plant that turns the lumber into birdhouses, the company that supplies the paint for the birdhouses, the logistics department that ensures the availability of supplies, the sales staff, the shipping department, the retailer, and the customer who ultimately purchases the birdhouse (see Figure 17.3). As we suggested earlier, not all of these organizational components may be under the direct control of the organization that heads the value chain.

Figure 17.3. Example of a supply chain.

graphics/17fig03.gif

Supply chains include many concepts, such as the following:

  • Collaborative planning and forecasting

  • Design collaboration

  • Coordinated manufacturing

  • Coordinated distribution management

These concepts represent such sophisticated processes that a number of large software companies, such as Manugistics and I2, have established a strong market in selling software to address them. Moreover, supply chain simulation software, such as that sold by Gymsym, provides users with the ability to model supply chains before they go into production. Even getting to this preliminary point depends upon a sound application integration infrastructure.

The supply chain determines the speed of operation and time to market, as well as the inventory an organization must carry at any given time. Supply chain management determines the overall cost of goods sold as well as customer service and satisfaction. Supply chain management relies on the planning and the control of the flow of goods and services, information, and money throughout the supply chain.

One more aspect to consider in understanding the need for supply chain integration involves the systems that enable these functions. These applications constitute possible data sources for a supply chain integration solution: In the enterprise domain, the inefficiencies spring from lack of process visibility among the functions. These functions are supported by applications, and a supply chain integration strives to tie these applications and processes together in a coherent, seamless whole to improve internal operations and minimize costs. These applications can be provided by a great variety of vendors and generally include complex, high-value engines for the generation of forecasts and plans for the management of operational processes.

Often, these applications are a world unto themselves without much exchange of information, even among applications from the same vendor. Carrying information from one system to another encompasses time-consuming and error-prone operations that may span a few enterprise functions or may be geographically scattered.

Having said all that, it becomes easier to understand and predict the problems that can arise from process isolation. For example, imagine that a discrepancy between demand forecast and actual orders placed is not communicated to production in an acceptable time: Either excess capacity or lost opportunity will result. Similar problems can be drawn, for example, from a production control that has poor communication with inventory control and procurement, or even between these functions.

In general, process isolation leads to:

  • Higher buffer inventories. Lack of timely visibility into procurement, demand, and production results in a higher operational risk; managers tend to stock up to reduce this risk.

  • Slower response to market demand. Information that is not reported in a timely manner or addressed by the different functions of the enterprise results in lower customer satisfaction, loss of opportunity (when the required goods cannot be delivered on time or at all), or excess capacity and high levels of finished inventory. Certainly, processes and systems have been devised to eliminate this problem, but the reality is that enterprises are still plagued by interprocess inefficiencies and miscommunications.

  • Forecasting methods have evolved into extremely sophisticated applications. However, these forecasts are only as good as the data they are fed and are only useful if their output becomes actionable information that is communicated to the appropriate enterprise function in time for action. The most accurate forecast is useless if its information is not acted upon in a timely manner.

  • The ERP revolution has promised near-perfect control of enterprise processes with efficient process integration. Yet, problems associated with process isolation still persist. ERP implementations place a significant burden on enterprises that chose to pursue them without a correspondingly high benefit. For example, several ERP implementations at different groups of the same firm have resulted not only in process silos, but also in application isolation, higher maintenance, and usage risks without a significant return on investment.

Often, enterprises spend significant effort and money in business process engineering, ERP systems, supply chain execution systems, supply chain planning systems, or other more-or-less integrated combinations thereof, only to realize that they still have difficulty decreasing their costs or affecting gains on the customer satisfaction front.

Forecasting and Supply Chains

Forecasts are approximated estimations of the behavior of the business environment not accurate representations. Although forecasts are used for long-range planning, they must be revised periodically with fresh data. This data can be historical (e.g., revise the forecast for the upcoming three months based on all the data available up to today) or forward looking. The former example has limited accuracy and puts enterprises in constant reactive mode in the face of changes in the business environment. The latter increases the accuracy of the forecasts and decreases the extent to which organizations react to their environment: Proactively responding to clients' needs or supplier peculiarities becomes a possibility. Clearly, access to the best data provides better forecasts; that's its link with application integration. Moreover, this information must come from multiple systems, both intra- and intercompany.

Although the ability to detect changes in the business environment provides for a considerable competitive advantage, it is necessary to make this information actionable and communicate it to the appropriate function for follow-up. This function may be internal to the enterprise or may be at a business partner (for example, change orders communicated to suppliers).

Although simple, this view of a value chain points to some needs in terms of information sharing that can have an adverse affect on the efficiency of the chain.

  • Deficiencies in responding to demand can have an amplified propagation through the chain. For example, suppliers vying to completely fulfill a forecasted demand by the customer will duplicate the orders.

  • Long lead times or infrequent availability may result in high inventory buildups.

  • High inventories may also exist as hedging against poor visibility into availability.

  • Slow response to demand (latency) spikes or dips generated by the primary customer may propagate through the chain and result in abnormally long lead times, a glut of capacity, high inventory, and so on.

The collection of data into the data core of a supply chain integration solution presents two distinct problems: a technical problem and a business problem. Both are related to the attractiveness of the solution to the participants in a chain what is compelling in a solution to create an incentive for partners to participate, above and beyond the value statements mentioned above. To this end it is important to ensure that participation in a supply chain integration solution be economical and that it introduce little to no risk into the participant's organization; that is, the collection mechanism should be quick and easy to install and simple to maintain. In a further stage, participants may decide to deploy their own solution in order to manage their respective relationships and chain.

A well-crafted supply chain integration and management system must create such a level of interdependency in a trading partnership that customers become locked in to such an extent that the cost of switching (not just the cost of goods but the cost of maintaining the relationship, as well) is so high that there is a permanent damper in the intent to switch to an alternate supplier. In a well-implemented supply chain integration environment, the flow of materials and monies is frictionless and the cost of altering the environment by substituting participants is potentially high. A well-crafted system will create an environment in which it is easy to add or delete a participant but difficult to replace an existing relationship.

Particularly in the extended enterprise domain, supply chain integration is not meant as:

  • Visibility into each and every process. This would result in redundant information that would be hard to maintain and update (or even make sense of). Tracking every bit of data leads to high installation and maintenance costs without a corresponding increase in value

  • Free-for-all visibility whereby any participant in the chain can view every bit of process data of every other participant. This adds little value to the chain and is bound to increase the potential for conflict and mistrust. A supplier may not be interested in sharing availability information about scarce or strategic components and may want to trade them outside of the supply chain integration framework. Likewise, a customer may not publish demand for a set of components that make up a strategic element of its product, and a supplier at a given tier may not wish to share availability information with potential competitors in the same tier.

  • Demand. Suppliers add great value from having visibility into their clients' forecasts as well as into their orders. This allows for more accurate and timely fulfillment, better operations planning, improved purchasing, and so on.

  • Supply. Visibility into supply (inventory and production capacity) adds great value to suppliers and to customers. The former can plan replenishment deliveries to the client according to this information, and the latter can plan their production and sales in the face of the known availability at the suppliers. This set is broken up into two subgroups:

    • Inventory

    • Production

  • Distribution. This data should satisfy the question, "Where is my order?" and it should allow for effective order tracking and monitoring of times of delivery. In simple terms and from an IT perspective a supply chain integration solution hinges on a few components:

    • Integration. This addresses a need that follows the nature of the problem: connecting a variety of systems across firewalls and integrating their data into a coherent whole (which is the concept behind this book).

    • Workflow management. This component solidifies the value of the solution. For example, the need to escalate and process an exception alert can only be addressed appropriately with some workflow management system. An exception for example, a discrepancy between forecasted and actual demand may trigger a workflow that spans the full value chain and may require a great deal of human intervention. Only a robust and configurable workflow management system can adequately address this need.

    • Data core. A data core wrapped in a standardization layer, which contains the set of data that represents the state of the supply chain, should represent data in a standardized format. It should also constitute the layer upon which more sophisticated uses are configured (e.g., workflow management and analytics).

    • Configurability. Supply chain integration solutions must be configurable to address these quirks of the business environment in which they are deployed. Solutions that have too rigidly defined workflows (or too loosely defined ones, for that matter) are unlikely to address the realities of most value chains.

Questions to Ask When Considering a Supply Chain Solution

When considering a supply chain integration solution, users must consider a series of issues that have deep, long-range impact on their operations and in the performance of the solution.

  1. Does the solution integrate only internal systems, or does it allow for partners to share their process information (e.g., information about actual demand, or changes in forecasted demand)? A true supply chain integration solution will allow for integration of information across the extended enterprise that is, across firewalls because as much value as it adds, integration of internal systems is an EAI problem, not a supply chain integration problem.

  2. How scalable is the solution? A well-designed solution will allow for the addition and removal of trading partners participating in the solution, as well as for the addition and removal of internal systems at one participant or another. The solution must be scalable enough to grow or contract with the chain.

  3. How strong is the transformation of the integration component? Given the multitude of systems involved in a chain that is three or four main tiers deep, this becomes an important consideration, and light transformation capabilities are unlikely to suffice.

  4. How invasive is the solution? A solution that requires writing wrappers, interfaces, or changes to existing systems is clearly not attractive. A solid solution must allow for the deployment of agents that carry the right adapters to the internal systems from which it will source data.

  5. To what extent does the workflow management component allow for human intervention? Full automation is a sure recipe for disaster: Following the very old law of "garbage in, garbage out," there are only so many "IF" and "ELSE" statements that can be programmed into a workflow manager before it runs out of computable options and botches a job. A well-crafted system will allow for rich, configurable human intervention. For example, it must allow for the creation of user interface screens in which users can view supporting, contextual information that allows them to make a decision, and exceptions not handled by the predefined rules must be brought to the appropriate user for decision and routing. Truly, a supply chain integration system is a robust way of marrying decision support with operational management, not just a decision support system or an operational system, and certainly not a reporting and analysis tool.

  6. Participants in a supply chain that are far removed from the main client (and owner of the hub) may also want to ask what is the compelling reason for them to participate in the solution. Take, for example, an automotive supply chain. When faced with the costs of installing and maintaining a new solution, a supplier of springs for car seats may have little incentive to participate in a supply chain integration solution because this business represents only a small fraction of their revenue. The costs and risks may not justify participation. On the other hand, participants who are closer to the end customer (say, the assembler of dashboards in the example above) are more compelled to participate for the opposite reasons. Thus, the manner in which data is sourced from the different participants in a supply chain integration solution must make participation economically compelling.

  7. The maintenance of visibility relationships poses an interesting problem to each participant in a supply chain. Generally, supply chain integration solutions are point to point. An enterprise must maintain several systems to exchange information with several clients, which introduces and amplifies the IT risks of having to deploy and maintain a great number of relationships. A solution to this problem at enterprises that maintain a great number of relationships is to pass deployment through a hub. The hub acts as a node in each chain and functions as the central point through which an enterprise manages the relationships it maintains with suppliers and clients.

What may catch your attention is that the questions above lead strongly to integration as the cornerstone of a supply chain integration solution. Although there is enough in a solid supply chain integration solution to differentiate it from an application integration solution, the reality is that supply chain integration without a very robust integration and transformation layer is simply not possible.[2]

[2] Mercator White Paper. 2002. "Supply Chain Visibility." www.mercator.com.



Next Generation Application Integration(c) From Simple Information to Web Services
Next Generation Application Integration: From Simple Information to Web Services
ISBN: 0201844567
EAN: 2147483647
Year: 2005
Pages: 220

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net