Uneven dwell time affects port processors, shipping lines and OEMs alike but is build-for-distribution a wish too far?
Even seasoned automotive logistics veterans are impressed at the scale of BLG Logistics’ flagship port facility at Bremerhaven, which handled nearly 2.15m vehicles in 2012 – another annual record, confirming Bremerhaven as one of the largest automobile ports in the world.
Yet look closely at the Bremerhaven operation and you’ll notice the seemingly slick movement of vehicles, from arrival at the port to departure on the vessels for export, is uneven. While the average dwell time might be short, variations mean some vehicles arrive and depart very quickly while others linger for days or even weeks.
About 95% of the vehicles arriving at Bremerhaven for export do so by rail, often on one of the 1,300 special railway wagons operated by BLG subsidiary BLG AutoRail. These wagons shuttle between Bremerhaven and the various carmakers’ assembly plants that feed it.
Vehicles often arrive at ports with little regards to the sailing schedule of vessels
There seems to be an elegant choreography when the vehicles arrive at Bremerhaven; they are unloaded, driven to a storage compound and loaded onto vessels, while trains loop back (sometimes with imports) to the assembly plants for the next batch. But the process masks the fact that the schedules for the vehicles arriving at the port actually have almost no regard for the timing or destination of their onward journeys.
According to Wolfgang Stöver, director of sales and marketing at BLG Automobile Logistics, incoming vehicles arrive completely unsorted by port of destination. “Vehicle production is optimised, the loading of the train is optimised, and the loading of the ship is optimised – but the car could arrive at Bremerhaven today, when the ship sailing to its intended destination actually left yesterday,” he says.
The best laid plans
In terms of information about those vehicles, BLG is far from operating in a vacuum. Stöver explains that the schedules supplied by carmakers offer considerable detail about the vehicles arriving over the next few days, along with information as to the country and port of destination.
“But while we know in advance the port of destination, the allocation of vehicles to specific vessels is only done when the cars are already in port,” he reveals. “The result is that we see an optimised supply chain stretching from the factories to the ports, which then comes to an abrupt halt when the vehicles arrive at the port, as the destination is served by a specific and contracted shipping line and the vessel’s departure may be four to six days later.”
"We can certainly conceive of a build-for-distribution production model, which takes into account the departure dates of the intended vessels...but it’s a question for the manufacturers" - Wolfgang Stöver, BLG
This could be even later in some cases, as some destination ports are only served by fortnightly sailings. The production planning of vehicle production doesn’t take this into account at all, says Stöver. Vehicles are produced to a manufacturing schedule then shipped to Bremerhaven as soon as possible to conserve scarce assembly plant compound capacity.
So could vehicle production planning take shipping frequencies and timetables into account, thereby eliminating the delays seen at the port of export? And could carmakers take into account other points of inefficiency in the downstream distribution chain? In theory, at least, some progress towards build-for-distribution seems possible. “We can certainly conceive of a build-for-distribution production model, which takes into account the departure dates of the intended vessels,” says Stöver. “But it’s a question that can only be answered by the manufacturers themselves. With the individual vessel sailings of shipping lines fixed – and known in advance – it certainly should be possible.”
While pointing to the problem, a logistics executive such as Stöver recognises that his OEM customers have their own scheduling priorities, choosing to optimise the flow of vehicles through the assembly plants with computer systems that use finely honed algorithms to juggle the competing imperatives of customer requirements, paint shop capacity, line balancing, and component supply.
But many are pointing to evidence that optimising vehicle production schedules does not result in an equally neat and ordered downstream logistics flow. In an industry driven by relentless cost-out pressures and intense competition, the costs and extra dwell times incurred by such single-minded, one-dimensional optimisation have to be taken into account.
“Our view remains very much that most automotive supply chains are ‘broken’ in the sense that they aren’t integrated enough, resulting in profitability seeping out of the system,” asserts Garth Parker, chairman of ProAct International, a specialist provider of multimodal supply chain planning, execution and cost management systems to automotive customers. “The whole end-to-end process must be managed holistically, as opposed to just moving inventory from one system to another, and then to another.”
What’s more, as the global automotive industry has evolved, those costs and the extra dwell time are mounting, which compounds the imbalances between production and distribution. This is partly because finished vehicles are being transported further than ever before. Not only have vehicle exports grown, or at least recovered from recent lows, on traditionally large shipping routes – such as from Japan and South Korea to North America, as well as from Europe to North America – but new export routes have developed. European OEMs export more to China and Africa; Japanese OEMs to Southeast Asia, and North American OEMs have grown links to Russia, the Middle East and Latin America. Trans-ocean shipment, once split between main trunk lines, is to some degree more fragmented than ever before.
Yet the extent to which carmakers’ import and export planning systems have recognised these shifts is debatable. In North America, for example, executives at Honda, for which larger export flows from the region are a new thing, admit that the company lack the systems integration to manage the chain (read more). But finished vehicle logistics executives say that even for many long established exporters, there is a tendency to miss the overall vehicle distribution picture while focusing on isolated events in the chain.
Carriers and ports take the strain
Talk to those close to the issue and there’s little doubt that those costs and extra dwell time mount up. Gabriela Stojicevic, corporate communications manager at shipping line Höegh Autoliners, says that given the costs of running a 27,000-tonne vessel with capacity for 7,800 cars, it is important to sail with as little empty space as possible. However, she says that very often cargo that is booked to be on a ship will not turn up at the port in time for the vessel’s departure.
“As we get closer to a port call, the customers will have booked their space on the specific vessel, and around a week in advance we will, to a large extent, know exactly what will be shipped out of that port,” says Stojicevic. “But we can never be 100% certain and in the final days before shipment, booked cargo might not always turn up at the terminal. Then it is often too late for us to bring in other cargo, resulting in empty space on the vessel.”
Don Asdell, president and chief executive of port processing provider Global Auto Processing Services, also points to issues that arise when vehicles of a specific type or intended downstream destination arrive suddenly at a port. “We call it ‘vessel bunching’,” he explains. “Instead of a steady stream of arrivals, we might see 5,000 vehicles arriving all at once, when we have the staffing and capacity to handle 500 a day. We much prefer to see volume spread out so that there are consistent and predictable dwell times. This ‘bunching’ strains the truckers, as well as the rail freight side of things.”
Many logistics companies are asking whether a build-for-distribution production and scheduling pattern is just a pipedream or if it something that carmakers could adopt.
“The industry is certainly starting to talk about it, which is a huge step forward,” says Richard Barker, managing director of Sovereign Business Integration, provider of the web-based Event Management Logistics supply chain management solution. “We’re talking to more and more people who want to explore with us how the finished vehicle supply chain can be better integrated, [including] shipping lines, port operators, and auto manufacturers.”
Some are doing more than just talking about it, too. The Volkswagen Group has discussed trials of geographic-based distribution at a number of plants in recent years. BMW has also at least been experimenting with the approach. “It’s a step-by-step process, market-by-market and plant-by-plant,” says Matthias Wellbrock, general manager for vehicle distribution at BMW. “With our new plant in China, we’ve just started another trial. The production planners in the plants are working together with our representatives in the plant dispatch areas.”
The results have been encouraging. “We see measurable benefits in reducing lead times, as well as inventory reduction at the plant dispatch areas,” he reports.
Wellbrock confirms that the market-by-market approach is designed to avoid adverse impacts elsewhere in the finished vehicle supply chain, especially in ports and in compounds for final distribution.
“We are not pushing the vehicles for the big markets, but are focusing on smaller ones, trying to build the cars right before the estimated departure date,” he explains. “For instance, if a vessel is leaving every second Monday and the lead time to the port is one day, we try to get the cars built on the Thursday and Friday before the sailing day.”
"We optimise distribution logistics post-manufacture, rather than as a total manufacturer-to-dealer process. It’s too big a challenge to [try and]reflect specific destinations" - Brian Mason, Toyota Logistics Services
“Everybody knows of the potential savings in transportation. At the same time, the restrictions in production and potential cost impacts are so huge that the build-for-distribution model is immediately compromised if it is impacted by bottlenecks or problems,” says Rudolf Luttmann, director of car and ro-ro handling in Europe at shipping line Mitsui OSK Bulk Shipping.
“We can all see the potential savings in lead time, inventory and optimal use of capacity, but so far the need for buffers in the system is still there,” he adds. “While the main reasons for these buffers are restrictions in production, and in transportation to the ports, at the same time we need to accept that sea transportation is subject to rough weather, and cannot be calculated to the minute.”
Arguably the world’s most efficient vehicle manufacturer, Toyota has yet to find build-for-distribution viable, at least for intercontinental exports. Manufacturing efficiencies often outrank logistics efficiencies, even if the cost is unnecessary dwell time. However, good planning processes do everything possible to smooth the route of finished vehicles, notes Brian Mason, national manager for strategic planning and communications at Toyota Logistics Services, based in Torrance, California.
Even as vessel loads are built up at the Japanese port of Tahara, Toyota’s main export port, Toyota personnel in the US can monitor them at VIN-level, he explains. An electronic notification of the manifest, again at VIN-level, is sent once the vessel sails, providing a 12-14 day planning window in respect of west coast destinations, and a 26-30 day window for Toyota’s east coast ports of Jacksonville and Newark.
However, Mason notes that while it is possible to allocate vehicles to dealers while on board a vessel, Toyota generally avoids doing so, except for Lexus vehicles. This is because the OEM prefers to allocate vehicles to dealers when it can supply a more accurate time of arrival. “Postponement allows us to manage our inventory better,” says Mason. “It helps the dealers if we allocate vehicles to them as late as possible, as then we know exactly what the dealer inventory is, and can make the best decision as to where to allocate vehicles.”
Mason is adamant that constructing manufacturing schedules to reflect downstream logistics plans – for example, batching by port – might well impede production efficiency, and so is avoided. Such an approach would be feasible at Toyota, as its regionally based national ordering system gives manufacturing planners access to the intended port of destination when constructing production schedules. But Mason doesn’t think it would be worth doing. “We aim to optimise distribution logistics post-manufacture, rather than as a total manufacturer-to-dealer process,” he explains. “It’s too big a challenge to go into the manufacturing plan and alter it to reflect specific destinations.”
So is that the final word? Perhaps not. Even Toyota, concedes Mason, can see the discrepancy between the downstream logistics efficiencies of North American-built cars and the flows of imported vehicles.
Domestically constructed vehicles, he explains, are generally allocated to a specific dealer before they’re even built – although there’s plenty of room for swapping and reallocating, either prior to manufacture or immediately after. Early allocation allows better planning of post-build logistics processes, making the best loads for trucks and scarce rail wagon capacity, with vehicles usually dispatched on the day of manufacture.
In short, domestic build is best described as a smooth, heijunka flow – which differs sharply from the lumpy arrival of 2,500 or so at a time, and which usually takes four days to process, or even longer for larger arrival batches.
The prize of build-for-distribution, then, remains there to be grasped – or to be abandoned, should carmakers decide that distribution is just one factor too far when it comes to planning production.