8/14/99

Trends in Automotive Supply Chains and Their
Implications for Enterprise Integration

From end-user to lower tier suppliers, changes are taking place which have implications for enterprise integration. Below we summarize these changes.

Business trend: Vehicle customization

Current practice in the auto industry is for dealers to finance and stock cars. Because of inventory costs, business is structured to make it attractive to a customer to purchase off the lot. If the vehicle is not on the lot, a centralized inventory control system is in place to find the vehicle on someone else's lot. As a last resort, an order is placed to the manufacturer. (This is a broad-brush description of a complicated process. The description does, however, provide an accurate overview of how the system presently works.)

One objective of the industry is to move toward mass customization, a system wherein buyers configure vehicles to their liking, and rely on an efficient supply chain to deliver the vehicle in a reasonable amount of time. The current system of mass production coupled with inventory stocking will not go away. Market conditions and production efficiencies will keep mass production an acceptable business model for the foreseeable future. Reliance on mass production, however, will diminish in favor of mass customization. The big question is how quickly the mass customization model will grow, and what percentage of production it will capture. Toyota has recently claimed that they will soon be able to produce a custom order within ten days. This may spur the US industry to faster mass customization than it had intended.

Inter-company implications for enterprise integration A forms/query system is needed that will let potential customers obtain information on the availability of uniquely configured products. To work, the system must link to dealers' inventory, OEM's production schedule and the OEM's production capability. In addition to availability, the system must generate pricing information on a much wider range of products than do existing pricing systems.

Mass customization will only work if the entire supply chain is characterized by an efficient two-way flow of information from top to bottom. When information flow is high, the required technology is standards based EDI. For members of the chain where information flow is low, EDI ß à forms/browser technology is needed.

If mass customization does indeed develop very quickly in the US automotive industry, then enterprise integration will have to bear the added burden of fast, large-scale implementation.

Intra-company implications for enterprise integration OEM's need pricing systems that can estimate the cost of an almost infinite number of vehicles which will be produced under conditions where prices vary with capacity in the supply chain. To accurately determine prices, estimating systems must provide feedback to the OEM, based on accurate information which passes up and down the supply chain.

In order for an OEM or a dealer to make believable promises about availability, all members of the supply chain, including the OEMs, will need better internal systems (ERP, MRP, order entry, PDM, etc.) than many of them currently have.

Business trend: Consolidation at the dealer level

The trend is toward fewer dealers, both in the form of direct sales by OEM's, and acquisition of weaker dealers by stronger ones.

Inter-company implications for enterprise integrationThis trend will actually make the information flow and connectivity problems simpler by lowering the number of entities that must communicate with each other.
Intra-company implications for enterprise integration With fewer nodes on the system, poor information from any one location will have larger overall effects.

Business trend: Lean service and aftermarket business

Just as with dealers, service and aftermarket sales have the problem of balancing inventory costs with customer service. Inventory is distributed in different parts of the supply chain - retail locations where sales take place, numerous OEM and independent distributors, and finally, at OEMs and independent aftermarket manufacturers. All members of this supply chain have an interest in reducing inventory while still maintaining a high level of customer satisfaction. To do so, they need to move toward a lean supply business model.

Inter-company implications for enterprise integration As with vehicle customization, this business trend raises the problems of on-demand manufacturing, inventory distribution and timely delivery. In addition, the aftermarket does not have the systems and standards in place to allow single-query checking for availability across different sources of supply. There are inconsistencies in part numbers, catalogues are not standardized, and query systems are proprietary.
Intra-company implications for enterprise integrationAll members of the supply chain already have computerized systems that do a reasonable job of identifying their stock. The primary issue for enterprise integration is how to successfully implement the new systems as (if) they arrive. The implementation challenge extends beyond technology, into the realm of business process. Members of the supply chain (some of whom compete with each other) must be willing to exchange information on parts availability

Business trend: Global sourcing:

Global sourcing is a process designed to support a trend in the automotive business to manufacture the same cars (or at least cars with many of the same components and subsystems) at different plants in different parts of the world. For this process to work, a variety of suppliers must be able to deliver the same part with the same specifications.
Inter-company implications for enterprise integrationOrdering will almost certainly take place via structured EDI transactions. The major integration issue is the use of X.12 in the US and EDIFACT elsewhere. Each supplier must be able to handle the system its customer is using. To complicate matters, its possible that a customer may find a need to use both standards, depending on the nature of its supply base. Further, each first tier company will have its own supply base using whatever system is in place at that level. Thus first tiers may be in the position of having to maintain one system for customers and another for suppliers.
One solution to this problem is for the automotive industry to adopt EDIFACT as its global standard. At another level, there is the matter of X.12/EDIFACT "harmonization". There is movement at both of these levels, but what will actually happen is unclear. At all points in the system where data movement is not clean, translation methods will have to be maintained. On the far horizon, XML may come to the rescue, but only if an appropriate level of standardization can be obtained. Even with the most optimistic time line, XML will not be in a position to replace X/12 or EDIFACT any time within the next few years.
Intra-company implications for enterprise integration Multiple data standards require the maintenance of multiple systems. The greater the integration of those systems into enterprise resource planning applications, the greater the maintenance burden.
An additional burden on integration and maintenance arises because in addition to non-standard data, systems must also account for non-standard parts. The solution is not as simple as the US going metric. For instance, within the metric system there are differences between countries in the number and pitch for standard screw threads.

Business trend: Closer coordination between OEMs and their suppliers

These ties are already very tight, as exemplified by the numerous successful JIT (just in time) systems that are in place. But however tightly coordinated, JIT is still a system in which the buyer orders from the seller. The next step after JIT is VMI (vendor managed inventory), a process in which sellers monitors their buyers' use patterns and makes the decision what to ship when. There is no evidence of an immediate mass movement toward VMI in the vehicle production function of the auto industry, but there are three reasons to think that its use may grow. First, it fits with the industry's trend toward outsourcing. (What is being outsourced is the function of determining production requirements.) Second, the industry already has experience using VMI for non-production material (MRO - maintenance, repair, and operations). Finally, several large OEMs (GM, Ford, Volkswagen) are working on manufacturing plants in which fabrication is done by different vendors within the factory, each responsible for its own subsystem. If this trend develops, VMI will of necessity develop with it.

Inter-company implications for enterprise integration First tier suppliers will need a direct window into their customers' production schedules and inventory levels. The necessary applications must be customized for each customer - supplier combination.
Intra-company implications for enterprise integration VMI is much more likely to develop in cases where a proven JIT relationship already exists, i.e. in cases where the customer and the seller already have the internal systems needed to supply accurate and current information. Thus we do not expect a shift from JIT to VMI to place a major burden on the companies involved.

Business trends: 1- First tier consolidation, and 2- continued reliance by OEMs on outsourcing for design and production

These trends are presented together because they are so highly related. OEMs in the automotive industry have greatly increased their reliance on outsourcing ever more complex components. Its uncertain whether the OEMs believe they have reached a satisfactory level of outsourcing, but the existing level is high and will remain so for the foreseeable future. As a result the first tier suppliers have been faced with a business challenge -- should they increase their reliance on their own supply chains, or should they acquire suppliers (and competitors) in order to improve both coordination and profits? More and more often these days, they are taking the latter course.
Inter-company implications for enterprise integrationThe enterprise integration challenge here has to do with product data. Along with outsourcing complex parts comes the need to exchange complex design information, with all its attendant problems of CAD translation, requirements for suppliers to maintain multiple systems, and product data management.
Intra-company implications for enterprise integration The internal problems mirror the external problems. Because of the huge consequences of working with incorrect design information, major resources must be devoted to assuring data quality. As the diversity of a first tier's customers increases, so do the resources needed to assure data quality control.

Business trend: First tiers will increase pressure on their suppliers for closer coordination

Along with the gains of increased design and supply work comes the responsibility to deliver more complex products on time and at a fixed price. To meet this objective first tier companies will press their suppliers for closer adherence to ever tighter delivery schedules.
Inter-company implications for enterprise integration There will be increased pressure on lower tier suppliers to do electronic data interchange. This pressure will be greatest (or at least most immediate) between first and second tier suppliers, but we do expect the need for electronic communication to ripple through the entire supply chain, putting a great deal of pressure on smaller suppliers. The first tiers will ask for a full range of interchange - standards based for the predictable, high volume information, and less structured and automated for the lower volume, less predictable data. In essence, the first tier companies will do to others as has been done to them.
Intra-company implications for enterprise integration There will be all of the well known requirements for integrated internal systems in support of external communication. An added complication, however, that many second and lower tier companies are relatively small, and do not have the technical expertise or the resources to implement complex systems.