Toyota Logistics Operation

December 29, 2009 - Tags:

Logistics is an extremely important component of the supply chain. It has two roles: (1) inbound logistics, which is responsible for transporting parts and materials from the tier 1 suppliers to the OEM plants; (2) outbound logistics, which is responsible for the distribution of vehicles from the assembly plants to the dealers. In this article we will examine both of these components.

Inbound Logistics

Inbound logistics encompasses two different operations: the first is the operation that transports parts from local suppliers to the local plants; the second is a separate operation, global inbound logistics, to transport parts from Japan to the North American and European plants. Because the inbound logistics operational models are very similar in both North America and Europe, we will explain only the North American operations. The local operation will be reviewed first, followed by an examination of the overseas operation.

Local Inbound Logistics

Toyota’s success in operating a lean supply chain requires that the parts be transported from the suppliers in an efficient and timely matter; therefore, Toyota establishes a partnership with a limited number of third-party logistics providers (3PLs) to deliver logistics services.

Toyota’s inbound logistics operation can best be described as a logistics network. The company organizes many of its suppliers into clusters based on geographic location. Parts are picked up from those suppliers by trucks on a “milk route” (i.e., a circuit in which a truck picks up multiple parts from various suppliers along the way), and then they are delivered to a regional crossdock. (Suppliers that are located close to the plants, however, ship parts direct.) At the cross-dock (a staging facility that is used to transfer parts), the parts are unloaded and staged for pickup and delivery to one of the Toyota plants. After the trucks arrive at the plant, the trailer is disconnected and parked in a numbered space in a staging lot. The trailers are not unloaded until the production progress triggers the need for the trailer to be unloaded. As discussed in Parts Ordering, all incoming parts orders and deliveries are synchronized to the production rate. Doing so ensures that the parts unloaded and delivered to the lineside workstations are just what is needed and just-in-time.

Network Logistics

The network logistics model enables Toyota to operate a very efficient and effective inbound logistics operation. Network logistics shows an example of a logistics network. The entities of the network are suppliers, cross-docks, and Toyota plants. The entities are connected by a continuous flow of trucks that move containers of parts inbound to the plants or move empty containers back to the suppliers. Plants include not only the assembly plants but also component plants that produce engines and transmissions. Toyota’s strategy is “small lots, frequent deliveries.” The ideal situation is for each supplier to ship parts every day to each plant. That course of events is where network design plays an important role.

The first step in network design is to analyze the location of the suppliers and identify clusters of them that are located in close proximity to one another. Next, a determination is made as to which cross-dock is located nearest to the suppliers. The idea behind this design is that one truck picks up parts from multiple suppliers in what is called a “milk route.” The truck then delivers the parts to the nearest cross-dock. The parts are unloaded and the corresponding empty containers are picked up and returned to the suppliers on the next run. The parts are then staged for pickup by trucks that are scheduled to deliver full truckloads of parts directly to each plant.

In the example in Network logistics, there are two clusters of suppliers, two crossdocks, and two plants. Parts from suppliers S1, S2, and S3 in cluster C1 are picked up by truck T1 and delivered to cross-dock CD1. Parts from suppliers S4, S5, and S6 in cluster C2 are picked up by truck T2 and delivered to crossdock CD2. Then, truck T3 picks up parts from CD1 and delivers them to plant P1. Also, truck T5 picks up parts from CD1 and delivers them to plant P2. Truck T4 picks up parts from CD2 and delivers them to plant T2. Finally, truck T6 picks up parts from CD2 and delivers them to plant P1. As you can see from this example, there are two milk run routes picking up from suppliers and four main routes from two cross-docks to two plants.

The advantages of the network logistics structure is that it enables Toyota to pick up from most suppliers on a daily basis while at the same time minimizing transportation costs.1 However, the network is extremely complex to operate and manage. Toyota’s size enables it to maintain control over the logistics network by partnering with 3PL companies. The logistics partners provide a dedicated fleet of trucks and drivers to operate Toyota’s logistics network. In addition, the entities in the network work closely with Toyota to design and plan routes. The shared transportation enables suppliers to receive small orders without increasing transportation costs.

Route Planning

Route planning is a key function that ensures efficient and effective operations. It is done once per month and is based on the next production month plan. Such planning is part of the production plan discussed in Production Scheduling and Operations and the parts ordering and forecast discussed in Parts Ordering. In these chapters heijunka was discussed as a method for leveling parts orders by day. Without a level production and parts plan, developing a daily logistics route schedule and repeating it for the whole month would not be feasible.

The creation of a logistics route plan to transport parts from hundreds of suppliers to multiple manufacturing plants is like the making of an airline schedule. The planner needs to know the locations of the suppliers, cross-docks, and manufacturing plants. Then he needs to know the number of packages or containers of parts to be picked up from each supplier each day and to which cross-dock and plant those materials are to be delivered. In addition, he needs to know how the containers of parts can be arranged and stacked within a truck. It is important to optimize the cubic space of each truck so that one doesn’t “ship air” and does avoid “blowouts”—in other words, one doesn’t ship partial loads or create a condition in which all of the containers cannot be loaded on a truck because of weight or volume restrictions (a “blowout”).

Another critical piece of information is the road routes and distances between all potential to/from destinations. A computer system is used to run simulations to create multiple route plans; they are then evaluated by logistics experts who select the optimal routes. The process is very complex, and numerous variables have to be considered (e.g., total miles, average miles per hour, number of trucks needed, number of drivers needed, and risks because of road conditions). The process used is an example of Toyota’s emphasis of combining the talents of human beings and the power of machines—Toyota doesn’t rely on the computer system alone.

Under normal conditions, the routes would not vary significantly from month to month. But in cases where there is a major change in production (either up or down) or if there are new suppliers, the change in routes could be drastic. That step would require more scrutiny to make sure that the plan is correct and that no error is in the simulation.

A Toyota logistics manager from the Princeton, Indiana, plant stated that for route planning, Toyota assumes a 50-mile-per-hour average speed of trucks and provides a desired route and travel time for deliveries. That time estimate is updated when snow or other weather conditions prevail. The planning thus shifts to this inclement weather route and associated lead time. Such detailed planning provides the plant with a good estimate of deliveries, and thus permits synchronization of parts flow with plant requirements.

Other situations that might arise include port-related unloading issues, border crossing delays because of tightened security checks, and strikes. In such cases, shipments are sometimes airlifted so that the flow is maintained. The capability to quickly react to impending crises enables alternate contingency plans to be generated and implemented to keep the parts flowing.

Pipeline Management

Toyota strives to operate an extremely lean supply chain, so it is critical for the plant production control personnel to understand the status of all parts in the pipeline. The “parts pipeline” is defined as all parts that have been ordered from a supplier and have not been unloaded at the receiving plant.

Toyota uses a variety of methods to track parts throughout the pipeline. The process starts with the parts order that is sent via Electronic Data Interchange (EDI), along with the kanban bar code label that the suppliers affix to the parts shipping container. Once the parts are shipped, the supplier sends an EDI Advanced Shipping Notice (ASN). The truck driver scans the kanban bar code label and identifies the truck onto which the parts are loaded. Once the truck arrives and is unloaded at the cross-dock, the parts status is changed to show arrival at the cross-dock. Again, as parts are loaded onto another truck bound for the plant, they are scanned and associated with the truck number. As the truck enters the gate at the plant, the parts status is updated to show that the parts are in the plant yard. The trailers remain in the yard until production progress dictates that they should be unloaded at the dock. As the parts are unloaded, each container is scanned to confirm the arrival at the plant. Pipeline data enable Toyota to have visibility into the parts pipeline. This pipeline database is especially important whenever there is a crisis situation such as parts shortage, short shipment, or transportation delay. It is thus clear that visibility plays a key role in the management of the inbound parts logistics process.

Some of the metrics used to monitor inbound logistics are percent of cubic capacity utilized, number of blowout loads, on-time pickup and delivery, and actual mileage versus plan.

Overseas Inbound Logistics

Overseas parts arriving from Japan are shipped via vessel to a port and then transported by railcar to the assembly plant. Once the railcar arrives at the assembly plant rail yard, the container is offloaded onto a truck and driven to the assembly dock. The trailers are parked in a large staging lot in a numbered space that can be used to locate the trailer.

One of the unique aspects of the parts flow from Japanese suppliers is the use of the vanning center. The vanning center is a consolidation point in Japan where parts are received from Japanese suppliers and packed for shipment to an overseas manufacturing plant. The vanning center operation is linked to the overseas parts shipping schedule described in Parts Ordering. At the vanning center, parts are packed into plastic trays. These trays are then arranged into groups to fit into a module for shipment. The modules are then loaded into containers for shipment by a container ship to the overseas port. Vanning packing process shows an example of the vanning packing process.

After the containers are loaded and shipped, it takes about four weeks for the containers to arrive at the overseas plant. After the containers arrive, they are staged in the lot outside the plant until needed. Normally, there are about three to five days of inventory in the lot. However, the containers are not unloaded Vanning packing process. Vanning packing process until the parts are needed for production. Similar to the method for local parts, overseas parts are unloaded based on the actual rate of production. Doing so keeps the parts inventory inside the plant to a minimum.

Because of the long lead time (six weeks) from when overseas parts are ordered until they are used for production, a risk arises that a parts shortage could require a part to be unloaded prior to its scheduled time in the production schedule. A parts shortage could occur for a variety of reasons such as an order error, excessive scrap, or a higher volume of vehicle order changes than expected. When a parts shortage occurs, a parts handling group at the plant must “tap” a container in the staging yard prior to its scheduled unload sequence. Tapping is a term that is used to describe the process for locating a container and unloading it out of sequence. There are several negative impacts on the plant when a container is tapped, two of which are excessive workload on the team members and a parts overflow. The reason why excessive workload results in a negative impact is fairly obvious, because it takes time to locate the container, move it to the dock, and unload the parts. However, it is not feasible to unload just the one part that is needed; the whole container must be unloaded. Doing so will create an overflow condition because there will not always be enough space to store these extra parts in the normal flow racks. That, in turn, creates an additional workload because these parts need to be handled multiple times and they could get misplaced.

The impact of variability at the plant is managed through appropriate use of buffer inventory for overseas logistics. That is facilitated by the visibility of the pipeline.

Long Lead Time Pipeline Management

Pipeline management is important for all parts, including local ones, but it is extremely important to monitor the pipeline for long lead time parts. Long lead time parts have a supply lead time of three to six weeks compared to less than two weeks for local parts. In addition, the vehicle order specifications are not frozen or finalized until about five to ten days prior to production. Therefore, if the dealers generate a high number of vehicle order specification changes, a parts shortage situation may be created because the long lead time parts ordered would be based on the forecast not the final order.

Toyota has developed a long lead time parts pipeline system to track changes to vehicle specifications on a daily basis and translate them into daily changes in parts. These changes are then compared to the parts pipeline inventory by day to highlight potential shortage and/or overflow conditions in advance. The results are presented in the form of graphs so that a visual representation is available of potential discrepancies that could result in a crisis situation. That information enables the parts manager to evaluate the situation, take an inventory of the parts in question, and, if necessary, place a special order for parts to be “air shipped” to avoid a production interruption. (Air shipments, however, are expensive and should be avoided unless a more expensive shutdown of production is imminent.)

To conclude this section on the inbound logistics operations at Toyota, it may be instructive to examine how the partnership between Toyota and its 3PL providers benefits not only to Toyota but also its partners.

Mutual Benefits from a Partnership

When Toyota partners with a supplier or a logistics provider, the benefits of the relationship are not for Toyota alone. The Transport Corporation of India (TCI) provides an example of how the partner can benefit. The company is a logistics provider in India that formed a joint venture with Toyota to deliver parts to Toyota, both imported (from the port) and sourced from more than 70 local suppliers. Initially, TCI learned how to better manage the delivery of auto parts from Toyota; since then, it has carried the best practices over to other manufacturers. For example, using lessons learned from Toyota, in the two-wheeler segment (namely, bikes—a very popular mode of transport in India), TCI redesigned delivery trucks to increase the number of vehicles carried from 50 to 58. TCI went on to change them into flexible trucks, then to use trailers to carry 85 bikes, and then to improve the trailers so that they could carry 110 bikes. TCI also added Global Positioning System (GPS) units to the trucks, so that manufacturers could directly track pending deliveries and plan their operations accordingly. In a country with poor roads choked with traffic and red tape at state borders, deliveries are liable to get stuck unpredictably; thus, tracking information can be vital to a manufacturer’s efficiency.

Starting out as a basic logistics company, TCI is now becoming a complex supply chain management provider. Other manufacturers have started listening to TCI. In some cases it has been asked to handle the entire inbound and outbound logistics. In other areas—for example, perishable products like chocolates—the company has made innovations such as linking the temperature of the truck with the GPS unit. In a one-year period from 2004 to 2005, the company’s worth increased from approximately $160 million to nearly $200 million.

Outbound Logistics

Outbound logistics is also known as product distribution, because the function of outbound logistics is to distribute the finished products from the OEM plants to the retailers. As discussed in Comprehensive Overview of Supply Chain, Toyota uses a different distribution flow in North America than in Europe. In addition, the relationship with the 3PL providers for outbound logistics differs from that for dedicated 3PL providers for inbound logistics. Although Toyota still considers outbound logistics providers to be its partners, those partners are not dedicated to Toyota because no one 3PL provider can control all transportation activities end to end from the plant to the dealer. Therefore, Toyota relies on common carriers, railroads and truck “car haulers,” to transport its vehicles from the assembly plants to the dealers.

Railroads ship many types of goods and raw material in addition to vehicles. They also ship vehicles from multiple manufacturers on the same trains. Trucking companies, like the railroads, ship vehicles for multiple manufacturers—in many cases, they mix vehicles from different manufacturers on the same truck.

North American Vehicle Distribution

North American vehicle distribution flow shows how vehicles move from the assembly plants through this distribution network in North America. After the vehicles are produced, they are shuttled into a marshaling yard. (Details of the marshaling yard operations will North American vehicle distribution flow. North American vehicle distribution flow be explained next.) Once processing is completed in the marshaling yard, the vehicles are shuttled to the staging area for shipment. There are two options for shipment of vehicles to the dealers. The first option is rail shipment, in which vehicles are loaded onto railcars, shipped to a railhead, and then loaded onto a truck for delivery to the dealers. The second option is for direct truckaway: vehicles are loaded onto trucks and delivered directly to dealers. Option 1 is used for dealers that are located a long distance from the plant, usually greater than 500 miles. They represent about 75 percent of the volume. Option 2 is used for dealers near to the plant—within two to three days’ travel time.

Toyota includes in its contracts with the trucking partners an on-time delivery objective of 48 hours from the time the vehicle is shuttled to the staging area to the time it is delivered to the dealership. A similar delivery standard does not exist for the railroads because so many variables, such as railcar switching time at rail yards, demand for empty railcars, congestion at final destination rail yards, etc., can impact rail shipment timing. Toyota also emphasizes quality by monitoring damage metrics for all of the trucking and rail partners. The company holds an annual meeting with all of its logistics providers to recognize the top performers in both on-time performance as well as quality performance. That recognition provides an incentive for the logistics providers to improve and also sets the benchmarks for future performance.

Marshaling Yard Operations

The marshaling yard operation is extremely important, as it ensures the efficient and timely delivery of vehicles to their final destination. Marshaling yard flow shows how vehicles flow through the marshaling yard. After the vehicles are produced in the assembly plant, they are shuttled to one of two areas. Vehicles that require installation of accessories go to accessory staging; all other vehicles are shuttled Marshaling yard flow. Marshaling yard flow directly to the rail or truck staging areas. Once the accessories are installed, these vehicles are then shuttled to rail or truck staging areas. Accessory installation usually takes from one to three days.

After vehicles are parked in the truckaway staging area, the trucking partner has the responsibility of loading the vehicles onto the truck and getting them delivered to dealers within the delivery standard. Toyota provides the trucking partner with a weekly forecast of vehicles by dealer. That information enables the trucking company to plan its operation to ensure an adequate supply of trucks and drivers. Although some fluctuation of deliveries by dealer will occur, the volume of deliveries to a cluster of dealers will remain relatively even. This is another example of the benefit of Toyota’s use of heijunka to smooth the production by destination to avoid spikes in deliveries.

Railcar loading is the responsibility of Toyota’s logistics division. Note that not all railcars are the same. There are two types of railcars for automobiles: “bilevel” and “tri-level.” Bi-levels are used to ship higher-height vehicles such as SUVs and pickups; tri-levels are used for smaller vehicles, including most cars. Some of the recent cars, especially the crossover models, are growing in height, which requires them to be shipped via bi-levels. That requirement results in increased transportation cost because rail shipment charges are based on a perrailcar cost. With a bi-level railcar, with a typical capacity of 10 vehicles (versus 15 for a tri-level), the cost could be as much as 50 percent higher. This is an example of a situation in which vehicle design can have a negative impact on the supply chain operations and costs.

Consider what happened at one of the Toyota plants: A major change for an existing model increased the height of the car by only a few inches, which resulted in the car being one inch too high to fit on a tri-level railcar. That extra height meant that the new model had to be shipped via bi-level instead of trilevel, resulting in an increased transportation cost.

The process for loading railcars is as follows:

  • Stage for rail shipment. Vehicles are staged in lanes by destination and railcar type. See Marshaling yard flow, which illustrates a rail staging area. Note how vehicles are parked in lanes by destination. In this example, there is only one lane per destination. If both bi-level and tri-level railcars were to be used to ship to each destination, there would be two lanes for each destination.
  • Prestage empty railcars. Empty railcars are shuttled by the rail company onto one or more rail spurs. These railcars are usually arranged in a string of six. See Marshaling yard flow for an illustration of two strings of six railcars.
  • Assign destinations to railcars . Once a new string of empty railcars is ready for loading, the dispatcher must decide which destinations are to be loaded. If possible, all six railcars should be loaded with vehicles for the same destination. The next best option is to combine destinations that will be on the same route; for example, vehicles shipped to New York and Boston would be picked up by an eastbound train. However, the dispatcher can load vehicles for destinations only if the staging lane is full. The dispatcher must take inventory of what is in the staging lanes to determine what can be loaded.
  • Load the vehicles. Vehicles are driven onto the railcars and scanned as they are loaded so the tracking system knows which vehicles are loaded into each railcar.
  • Release the railcar for shipment. Once the complete string of railcars is loaded, the dispatcher contacts the rail company to pull the full load out of the yard and replace it with an empty string.

This process appears to be a straightforward one; however, it does not always operate effectively. The following situation that occurred at one of Toyota’s plants in 2003 provides some insight into how conflicting objectives can result in a negative impact on downstream operations.

At one of Toyota’s North American assembly plants during 2003, there was a great deal of tension between the plant marshaling yard management and the rail company local management. The main concern of the yard management was that they were under strict orders from the plant manager to load railcars as quickly as possible; that sense of urgency ensured that vehicles were moved out with minimal delay so that delivery time to the dealers would be kept short. The yard managers suspected that the rail company personnel did not share the same sense of urgency, because each day when the plant manager passed by the rail yard on his way to work, he noticed that several railcars that were loaded the previous day were still parked along the fence. When the yard managers discussed the matter with the rail yard management, the reason given was that the rail yard personnel took several hours to shuttle railcars around and match them up with the proper outbound trains.

This conflict continued for several months; there did not seem to be any resolution. To make matters worse, the plant was planning a major increase in production within a year, so if this problem could not be resolved, the rail yard would need to be expanded at a cost of several millions of dollars.

Toyota management decided that the situation was critical, and so they dispatched an independent team from other Toyota locations. Team members could objectively study and evaluate the situation, recommend countermeasures, and attempt to find a compromise.

They arrived on site and met with the marshaling yard management to request a detailed explanation of the railcar loading process. After taking a walking tour of the yard, they were directed to a whiteboard that showed the number of vehicles ready for shipment by destination. That board was the dispatcher’s “bible” and indicated visually the actual inventory in the staging lanes available to be shipped by destination. It was continuously updated during the day, and as soon as there were enough vehicles to load six railcars, the dispatcher called for an empty string of railcars and proceeded to load the vehicles. Because the rail yard’s focus was to move the vehicles through the yard quickly, the sequence of the loading was based on FIFO, or first in, first out. Rail yard personnel were very proud of their operation because one of their key metrics was “days of yard inventory” and they were averaging less than one day.

Next, the team met with the rail yard management (the rail yard is owned by the rail company; these managers did not work for Toyota). The rail yard managers explained that the reason that some of the railcars were still sitting in the yard a day after they were loaded was because each string of six railcars had to be disassembled and rearranged into strings to connect to either the eastbound or westbound trains. Also, the vehicles were not always loaded in the proper sequence. The constant shuttling of railcars was a burden, and they indicated that they would need to add more engines and people to handle the increased volume.

The team asked, “Is there anything that Toyota could do to reduce your workload and increase the throughput?” The response was surprisingly quick. Rail yard managers suggested two changes to the process for loading railcars:

  1. Load only vehicles with western destinations in the morning so that they can be shipped on the westbound train that departs at 1 p.m. Then, load only vehicles with eastern destinations in the afternoon so they can be shipped on the eastbound train at 9 p.m.
  2. Load a string of six railcars based on the sequence in which the destinations will be delivered. For example, if there are railcars going west to Los Angeles and there are both Denver and L.A. railcars on the string, place the L.A. railcars at the front and Denver railcars at the rear. With that arrangement, when the train arrives at Denver on the way to L.A., the Denver railcars can simply be disconnected without affecting the L.A. railcars.

If the Toyota yard managers would agree to this change, then that decision would reduce the rail yard’s need to disassemble and rearrange each string of railcars.

The team went back to the Toyota marshaling yard managers and explained that loading the railcars without regard to destination was causing extra work for the rail yard personnel. In fact, it was actually delaying the shipment to dealers because there were only two train departures per day: an eastbound train departing at 9 p.m. and a westbound train departing at 1 p.m. So, if vehicles for westbound destinations were loaded in the afternoon, they would not depart until the next day, and if vehicles for eastbound destinations were loaded in the morning, they would not depart until 9 p.m.

The response from the Toyota yard managers was lukewarm. Although they could understand the potential benefit, their concern was that such a change would result in an increase in yard inventory and would most likely require Toyota to expand the yard capacity to accommodate the future increase in volume. The team was concerned that an impasse seemed to have been reached, so they made a proposal: create a simulation based on the last month’s loading data to show what would have happened if the railcars were loaded based on the process recommended by the rail yard manager. The simulation showed that the same number of railcars would be loaded each day and that there would not be a significant increase in yard inventory.

The next step was to do a pilot for two months to guarantee that there would not be any operational problems. The pilot was successful, and the new loading process was implemented. The increased throughput was great enough to avoid any expansion of either the marshaling yard or the rail yard. The benefit was not only a cost avoidance of millions of dollars but also a shortening of the vehicle delivery time because the railcars did not stagnate in the rail yard.

This example illustrates that it is imperative for logistics managers to consider the impact on downstream operations when they are establishing or modifying processes.

Tracking Progress

Tracking the shipment of vehicles from the plant all the way to the dealer is crucial. Toyota has created a tracking system that receives input from the rail companies on a real-time basis that reports the progress of each railcar. It also gets input from the trucking companies when vehicles are delivered to the dealers.

Toyota uses this information internally to monitor the distribution progress throughout the logistics network. It also provides an estimated time of arrival (ETA) to the dealers. The ETA calculation is based on the date and time of the final quality assurance (FQA). The FQA is the point in the marshaling yard process at which a vehicle is ready for shipment. The calculation is a follows:

ETA = FQA + estimated delivery elapse time

The estimated delivery elapse time is based on the recent actual history for each route.

If a vehicle has not been produced, the ETA is still calculated; the FQA date must also be estimated. Because many variables can impact the actual transportation time, Toyota provides the dealers with an estimated threeto five-day arrival window instead of an exact date. That arrival window is updated as the vehicles get closer to delivery. The ETA information is used by the dealer to provide its customers with expected delivery time.

Distribution Flow in Europe

Toyota’s distribution flow in Europe is different than its flow in North America because the dealers in Europe do not have enough space to maintain a large inventory of vehicles. In fact, many dealers operate in an urban area and have room for only a few showroom vehicles. Europe vehicle distribution flow shows the normal flow in Europe vehicle distribution flow. Europe vehicle distribution flow Europe. As in the North American flow, the vehicles are released from the assembly plant and shuttled to the marshaling yard. In Europe, there is no additional processing in the marshaling yard. Instead, the vehicles are shuttled directly to the staging area for shipment via truck to one of the regional hubs. The hubs are where the installation of accessories and the application of the price label take place. Once the vehicle is sold by a dealer, the hub is notified to prepare the vehicle for shipment and install the accessories. Then the vehicle is shipped via truck to the dealer.

One of the other differences in Europe is that most transportation is via truck, not rail. Nevertheless, there is some limited use of rail for long-distance routes, and of course ferries or ships are used to cross waterways.

Another difference is the method for calculating the ETA. The ETA is based on the FQA date and time from the hub, not the marshaling yard at the factory. Also, the ETA is a promise date, not a delivery window. The promise date reflects the latest estimated arrival date.

Toyota’s distribution processes in North America and Europe are just one example of how Toyota has the agility to adapt its processes to different environments around the world.

The paragraphs that follow give an example of how other OEMs value the importance of logistics as a key component of the supply chain.

Outbound Logistics at Ford and General Motors

As Ford and General Motors try to reduce the cost and increase the reliability of delivery in their outbound logistics, they both have tried using specialists to manage their operations. Ford created an alliance with UPS Logistics Group (a subsidiary of United Parcel Service) in 2000 and GM formed a joint venture with the logistics and freight giant Con-way (then called CNF).

UPS did not actually transport cars for Ford but provided their logistics expertise for management and tracking of deliveries. Ford’s distribution network, based on a hub-and-spoke system with rail and ground transportation, utilized 14 different carriers that worked independently on their own lane optimization. UPS reengineered the network for simultaneous delivery planning across all the carriers, enabling optimum analysis of vehicle demand, assignment of sourcing, and scheduling of loads and delivery for all Ford dealers. According to a joint press release by Ford and UPS in 2001,3 within one year they had shaved off four days of delivery time. They saved $1 billion in inventory and more than $125 million in annual inventory carrying costs for Ford.

General Motors went a step further to create the largest such venture in the automotive industry, outsourcing its entire inbound and outbound logistics operation worldwide to a joint venture formed with Con-way.4 This new company, called Vector, was responsible for managing shipment of parts to plants, vehicles to dealers, and aftermarket parts and materials. The idea was to use a common information technology system to obtain seamless visibility of materials, parts, and vehicles moving through GM’s supply chain worldwide. By July 2002, GM had reduced logistics costs by 10 percent and reduced time for delivery from 15 days (in 1999) to about nine days.

Although this experiment was successful, GM ended the joint venture in 2006 by buying out Con-way’s 85 percent share in the company. The reason cited was a response to changing fortunes to support GM’s turnaround activities in North America.6

Reflection Points

Toyota’s logistics operation is an excellent example of where Toyota’s organizational learning practices are extended beyond the enterprise to include the 3PL partners. The continuous sharing of the Toyota Way principles and use of v4L with their partners ensures that parts and vehicles are transported in an effective and timely manner.

  • Velocity. Parts inflow, production rate at the assembly plant, and rail departures are synchronized to ensure heijunka across the supply chain. This process enables a steady velocity to be maintained.
  • Variety across vehicles affects vehicle height, which in turn affects loading efficiency in the railcar (bi-level or tri-level) and thus affects associated costs. Destination of delivery affects vehicle options and flow time and is in turn linked to parts inflow and finished goods outflow. Dealers have some limited flexibility to swap vehicle orders with other dealers’ orders in the pipeline.
  • Variability is managed by combining pickups across suppliers to create milk routes and the use of cross-docks. Tapping parts compensates for variability by using pipeline inventory.
  • Visibility across the pipeline of parts inflow plays a key role in maintaining a lean system, particularly for long lead time parts. Scanners, ASNs, and the like enable visibility across the supply chain. Continuous monitoring of outbound flows from the plant all the way to the dealers provides visibility for dealers.

Toyota takes many steps to foster learning and continuous improvement in its logistics operations. These include the following:

  • Create awareness. Toyota monitors the pipeline both to obtain early signals of problems that might arise and to take corrective action based on pipeline information. Progress of shipments is tracked with precise calculation.
  • Make action protocols. Trade-offs that must be considered when designing the logistics systems (e.g., on-time pickup, capacity utilized, and total actual mileage) are clearly defined. Contingency plans are drawn with care to address exceptions that arise from time to time.
  • Generate system-level awareness. The information system makes the supply pipeline visible. System-level exceptions are studied to isolate major issues, and those concerns are addressed on a priority basis.
  • Practice “go-and-see.” The highest-ranking managers go to the site with the logistics problem so they can see firsthand. These managers take a systems approach to solving problems, considering the impact both upstream and downstream.
  • Make deviations visible. Toyota levels the workload with milk runs. That step makes deviations obvious and immediately visible.