1、 外文翻译 原文 Delivering the 3-day carthe strategic implications for automotive logistics operations Material source： Journal of Purchasing Hall, 1993), but this research for the first time seriously questions the viability of the make-to-forecast and sell-from-stock approach. To date in Europe, only hal
2、f of all volume vehicles are built to order, the remainder are sold from stock. In the UK, this percentage is even lower at 32% (Kiff, 1997;Williams, 1999). Yet the cost of holding inventory at the most expensive point in the supply chain (Fisher, 1997), and discounting needed to sell these vehicles
3、 are costing the industry dearly. The predicted nancial benets of building vehicles to ordermarkets (Lapidus, 2000;Roland Berger, 2000). It is therefore hardly surprising that most manufacturers are working, more or less openly, on their build-to-order initiatives. Examples include Volvo with their
4、21-day order-to-delivery target (Hertz et al., 2001), BMW with their 10-day car, Ford with their 15-day Order Fulfillment project, and Renault and Nissan 14-day car within their Projet Nouvelle Distribution (PND) and SCOPE projects, respectively. The transition from forecast-driven to customer-drive
5、n vehicle supply will have wide ramications for the whole supply chain. Not only will suppliers have to be more exible, it also puts a question mark on current global sourcing strategies. Whilst build-to-order has been discussed on a generic level (Holweg and Jones, 2001;Holweg and Pil, 2001), this
6、paper will focus specifically on the logistics aspects of build-to-order. As the connecting element in the supply chain (Bowersox and Closs, 1996), logistics plays a crucial role within a build-to-order system, which is best highlighted in the case of Renault. At the outset in 1999, the company had
7、bold plans with PND, aiming to reduce the order-to-delivery cycle to 14 days. Since then, Renault has cut inventory by 21% and reduced expenditure byh513 million in the rst 6 months of 2001through stock reduction and supply chain alignment to customer demand. Speaking at the Frankfurt Motor Show in
8、2001, executive vice president Pierre-Alain de Smedt said: Were economising in logistics, reducing stocks, shortening delivery times and simplifying the entire delivery chain. Renault hoped that the next phase of the programme would reduce expenditure on procurement, operations and distribution by a
9、 total ofh3 billion by 2003. In 2002 Renault had to revise their target from 14-day order-to-delivery to 21days (de Saint-Seine, 2002), as only 21% of vehicles were received within the projected 14 days. According to Renault, the main reason for this failure was the unreliability and inexibility of
10、the logistics operations to cope with the short delivery times needed to support a 14-day car. This reinforces the fact that reducing manufacturing lead-time is of little impact, if the logistics lead-times are not addressed (Daugherty and Pittman, 1995). In this paper we aim to investigate the inhi
11、bitors of build-to-order in current auto logistics operations, how these could be overcome, and what cost and environmental implications would result from adopting a build-to-order strategy. Rather than focussing on a single type of logistics operations in isolation, we aim at providing an overview
12、of the main issues and implications of the major logistics operations in the automotive supply chain. This also allows for comparison of the issues within the three main logistics operations: inbound operations between rst tier suppliers and assembly plants, vehicle distribution logistics between fa
13、ctories and dealers, and the deep and short-sea vehicle shipping operations. The latter seems less obvious, yet due to the geographical dispersion and the low transportation cost per vehicle (e.g. $700 from Japan to the UK, $250 US East Coast to Europe), sea transportation is an integral part of the
14、 system. Across the globe, nearly 7 million new cars are transported via the sea every year (Beresford et al.,2000). Even within Europe, short-sea car transportation is very common due to the low cost involved. The research approach is based on case studies of inbound, outbound and a sea transport o
15、perations within the UK. Rather than using surveys, it was felt that a case-based research strategy would be more suitable to provide the necessary in-depth understanding of the current process and issues within these logistics schemes. Within each case study, a detailed 12 day process mapping works
16、hop was conducted, following the Learning to See Value Stream Mapping methodology (Rother and Shook, 1998), to provide an in-depth understanding of the process. A Learning to See map captures the entire order fullment process, comprising of both information and material ow, on a single piece of brow
17、n paper or a wall-chart, which is why it is often also referred to as Big Picture Mapping. Furthermore, a series of 26 semi-structured interviews was conducted with operations and planning staff at 6 logistics companies and 4 vehicle manufacturers in order to cover softer issues and perceptions of t
18、he key problem areas in the system. Also, where available, secondary performance data was collected. Overall, this multi-method approach allowed for data triangulation (Jick, 1989;Mentzner and Flint, 1997). A main concern with case-based research, since its origin, has been the generalisability of t
19、he ndings derived from single cases (e.g.Gummesson, 2000; Silverman, 2000). In order to mitigate bias in the case selection and to ensure general validity of the ndings, multiple cases studies were conducted. In detail, three cases each of inbound and outbound operations have been analysed, plus one
20、 vehicle import/export operations of one of the worlds leading car shipping companies. The cases comprised of three inbound collection schemes delivering predominantly into UK car assembly plants, three vehicle delivery operations from UK car plants into the UK and international markets and one vehi
21、cle import and export operation. Based on the process maps and secondary performance data from all cases across the three logistics operations analysed, this section gives a brief explanation of each process in order to provide a background for the overall discussion. The focus will be upon the gene
22、ric process and key ndings only. A more detailed discussion of the inbound and outbound operations, respectively, can be found inHolweg and Miemczyk (2002), andMiemczyk and Holweg (2002). The main function of the inbound logistics process is to collect goods from suppliers, to re-assemble these into
23、 delivery loads per time slot for each of the assembly plants, and to deliver these at minimal cost exactly when they are needed. Cost is determined mainly by three factors, inventory in the system, load efciency, and vehicle mileage on both collection and delivery runs. The process described inFig.
24、 1shows a generic supplier collection scheme based on UK suppliers and international deliveries feeding into a UK vehicle manufacturer. The processes were similar across all cases, with regard to key process steps. Small differences in the operational performance relating to lead-times and efciencie
25、s were found, but did not add to the comparative analysis. In all cases, which cover supplier collection of 470 suppliers and deliveries into 7 assembly plants, a third party logistics company organised collections from a selection of suppliers. These were delivered into the vehicle manufacturer (VM
26、) either through a milkrun direct to the VM plant, a milkrun through a cross-dock warehouse, or direct from the supplier to the VM plant. Collections from suppliers were made between once a week and more than daily, depending on bulk and parts volume. On average, a typical VM requires daily or more
27、collections from around 50% of its suppliers, with the rest of the suppliers being collected less frequently. This is based on current demand variability, schedule lead times and schedule reliability. Each of these would change if 3-day cars were a major proportion of VM Production. In contrast to t
28、he similarities of the physical operations, the main differences between the schemes relate to the information ow. Across the cases, the collection requirements for the logistics companies were either transmitted directly from the VM (and concurrently to the supplier), transmitted from the supplier,
29、 or transmitted through the logistics company to the supplier. In the rst case the information direct from the VM allowed a rapid response for collection, as the delivery and collection schedule could be amended within hours, in case of last minute changes to the build schedule. In the last case, th
30、e collection scheme requires a 2-week change interval, as land-based mail is still used as means of communication for the shipping schedule. Upon questioning this practice, legal requirements (for paper copies, rather than electronic documents) were given as reason. Although actual transit times wer
31、e very similar between 24 and 36 h, the more rapid information ow obviously allows for quicker response to changes in the VMs material requirements. In addition, high- variety packaging and the use of cardboard in general was seen as adversely affecting load efciency. The number of containers ranges
32、 from 270 (with 10% standard sizes) to 120 (with 60% standard sizes) across the cases. In conclusion, all cases show short overall lead-times, yet, due to the differences in the information processing, have very different abilities to react to changes of the vehicle manufacturer. It should be noted
33、that these changes generally only concern the quantities and types of parts collected, rather than the ac tual routing of the collection scheme itself, which would take much longerto modify in all three cases. 译文 “三天汽车”运递模式 一种对汽车物流操作业务具有战略意义的模式 资料来源 ： 采购与供应管理月刊 作者： 马提亚 霍尔维格 ，琼 迈因奇克 汽车产业内现行的 “ 挤压存货 ”
34、 的供应方式是让客户大量地下单订购现有的库存产品。但这种供应方式现在愈来愈显示出它的弊端。市场上不断升高的汽车存 货量和售卖汽车时的必要贴现都日益增加了汽车厂商的成本和压力。汽车厂商不得不重新思考他们的订单完成策略，而渐渐青睐于建立趋向零库存的“ 按订单制造 的生产模式 ” 体系。然而对于一家汽车生产商，更加快捷地完成采购订单不但需要灵活高效的零部件供应和整车组装的配合，而且还要为汽车供应链上所有物流操作建立广泛的销售渠道。基于 “ 三天汽车 ” 调查项目的研究结果，本文比较了入境、出境和海运运输物流的含义，旨在开发一种适用于将来汽车物流操作的战略框架。 在第二个汽车百年开端之际，全球汽车产业
35、处在了一个重要的转型期 按理说应该是处在 了它最有意思的一个阶段。在亨利福特那个时代，涌现的基本上是多品种的大规模生产，而精益生产的概念还只是处在被采纳的初期，今天全球的汽车产业在如何重获盈利能力的问题上面对着非常不确定的因素。众所周知，汽车产业的盈利平均少于 息税前利润 （ EBIT）的 4%。全球产能过剩和汽车存货量升高都威胁着整个产业。汽车制造商试图通过一系列（经常结果难以预测）的全球 兼并与收购 期望能以平台和零部件共享的方式取得更好的规模经济效益。与此同时，先前的那些核心竞争力产品，包括零部件或组件组装业务都 外包 给了一些比较 大的 第一 阶供应商 其中一些工厂已经在销售额和规模上
36、超过了他们的汽车制造商客户。更远的不确定性来源于欧洲的汽车寿命末期法令，它要求生产商到 2015 年底回收利用所有汽车的 95%部件。如果该法案以它现在的样子实施起来，那么它将在整个汽车供应链中带起一波零部件和各种物质回收利用的逆流。 总之， “ 产业中的产业 ” ，德鲁克（ 1946 年）曾用它来指汽车产业，现在在其运行中寻觅着下一个有竞争力优势的资源，这就是为什么近几年中 “ 按订单制造 ” 受到这么多的关注的原因。 按订单而 不是按预测生产汽车的并不是一个新想法。它已经被讨论将近十年了 (Delbridge and Oliver,1991;Hall,1993), 但是一开始这个研究是用来
37、质疑 “ 按预测生产的方式 ” 和 “ 按库存销售的方式 ” 的。至今在欧洲，只有半数的汽车是按照订单来制造的，其余的都是按库存来销售的。在英国，这个比率甚至更低，为 32% (Kiff,1997;Williams,1999)。然而在供应链中库存持有的开销占了最大的比重 (Fisher,1997),而售卖汽车时的必要贴现也消耗了制造商大笔的费用。在全球市场，预计按订单 量来制造汽车所获的财务收益是每辆 500-1500美元(Lapidus,2000;RolandBerger,2000)。这就难怪大多数的生产商都会起初几乎都按订单来生产汽车。这样的例子包括：沃尔沃的 “21天订单生产期后的交货目
38、标 “模式 (Hertz et al.,2001)，宝马的 “ 10天汽车 ” 模式，福特的 “15天订单实施项目 ”模式，雷诺和尼桑各自在他们的 “ 新的销售计划 ” （ PND）和 “ SCOPE 计划 “中的 “14天汽车 “模式。 汽车供应从预测为导向到以顾客为导向的转变将为供应链上所有物流操作建立广泛的销售渠道。这 不仅促使供应商必须更加灵活处事，而且它也在现行的全球资源策略上打了个问号。同时 “ 按订单制造 的生产模式 ” 在属层中也被讨论过 (Holweg and Jones,2001; Holweg and Pil,2001), 这篇论文将集中专注在物流方面的 “ 按订单制造
39、的生产模式 ” 上。 作为供应链中的连接单元 (Bowersox and Closs,1996)，物流在 “ 按订单制造的生产模式 ” 扮演了关键的角色，这在雷诺汽车例子里最为突出。 1999年初，雷诺公司采用了 PND 这个大胆的计划，该计划旨在将 “ 按订单制造 的 ” 周期缩短到 14天。从那时开始，雷诺公司已经砍掉了 21%的库存，并通过缩减库存和让供应链直接对准客户的需要在 2001年的 6个月里减少了 5亿欧元的开销。在2001年法兰克福汽车展上，雷诺的行政副总 Pierre-AlaindeSmedts 发言道： “ 我们在物流方面开源节流，缩小库存，缩短运递时间，简化整个运递链。
40、 ” 雷诺公司希望下一阶段的规划将在采购，运营和销售方面把总开销到 2003年为止减少到 30亿欧元。 在 2002年雷诺公司不得不把他们的目标从 14天下订单后运递增加到 21天(de Saint-Seine,2002), 因为只 有 21%的汽车达到了在计划中 14天的目标。据雷诺公司透露，这次失败的主要原因是物流操作在处理 “ 14天汽车 ” 的短时间运递中存在着不可靠性和不灵活性。这说明缩短生产商的交货期还是会有些影响的，如果物流方面时间没有安排好的话。 (Daugherty and Pittman,1995)。在本文中，我们旨在找出现行的汽车物流操作中抑制 “ 按订单制造 的生产模式
41、 ” 的不利因素，研究出如何克服它们的对策，并调查出运用 “ 按 订单制造 ” 的策略会产生的费用和它对环境影响。 我们不是孤立地关注单个的物流操作，而是旨在提供关于主要问题和涉及汽车供应链中重要物流操作的总看法。这也使得我们必须比较三个主要物流操作中的问题：在第一阶供应商和组装工厂之间的入境运输物流，在工厂和销售商之间的汽车销售物流，远海和近海中的汽车运输物流。最后一个物流操作看似不太显著，然而由于地理分散和每辆车的低成本运输（比如从日本运到英国只要 700美元，从美国东海岸运到欧洲只要 250美元），海运应该作为该体系中不可分割的一部分。每年几乎有 700万辆新车通过海运到达世界各地 (B
42、eresford et al.,2000)。由于运费低廉，甚至在欧洲境内汽车的近海运输也是非常普遍的 。 本文的研究方法是基于英国的入境、出境和跨海运输操作的案例研究。相比于调查研究， 我们 感觉基于案例的研究策略会更适合 于 深层次剖析这些物流主题的现行流程和问题。根据 “ 学会看 ” 价值流程图的方法论 (Rother and Shook, 1998)，在每个案例研究中都导 进 一个 细致 的 需要 1-2 天 时间运算 的过程图工作室来深层次剖析物流流程。一张 “ 学会看 ” 价值流程图能通过对比信息流和物流 来获取 整个订单的实现过程。而 之后流程图可以 打在单张牛皮纸或者是挂图上 作
43、为参考 ，这就是它被经常称为 “ 大图片制图 ” 的原因。 另外，对 6 家物流公 司和 4 家汽车制造公司 中的 操作规划员的 26 次半结构访问为的是得到 物流 体系中核心问题领域的 “ 简化后 ” 的问题和看法。还有如果有可能的话， 还 能收集到辅助绩效数据。总之，数据的三角测量方法也适合于这个多元方法探究 (Jick, 1989; Mentzner and Flint, 1997)。 自以案例为基础的研究 问世后 ，人们主要是关注它概括单个案例研究结果的能力 (e.g. Gummesson, 2000;Silverman, 2000)。为了降低案例选择中存在的人为偏见因素，保证研究结果
44、的总效力，我们采用了多案例研究 的方式。更细得说，我们都会对三个案例中的每个 出入 境操作进行分析，另外还会对 全球 领先的汽车运输公司的汽车 出 口 /进口进行研究。这些案例包括了主要把汽车 运送 到英国汽车组装厂的三个入境 采购 主题，将汽车从英国汽车工厂运到英国和国际市场的三个汽车物流操作，和一个汽车进出口操作。 基于从三种物流操作的所有案例研究中得到的流程图和辅助绩效数据，这部分会给出关于所有流程的简要概述， 从而 为整个讨论提供背景。 我们的 关注点只会落在总流程和关键结果上。而对出入境操作的更多讨论请参见霍尔维格和 迈因奇克 (2002)， 和 迈因奇克 和 霍尔维 格 (2002
45、)。 入境物流流程的主要功能是从各供应商那 采购 产品零部件，重新 对 它们进行整理并每隔一段时间为每个组装工厂进行装载，为节约成本只在工厂需要时将它们运送给工厂。期间费用主要取决于三因素，物流体系中的库存，装载效率，和包括采集和运送两方面的总里程汽车运费。如表 1 所示，这个流程展现了一个供应商采购的总主题，这个主题是基于多家英国供应商和承运商将零部件供给一家英国汽车生产商的实例得出的。 所有案例中的流程都有着类似的关键流程步骤。在操作表现上的些许不同在于我们 发现的 那些运送间隔时间和运送效率，但这两点都 不会纳入比较性分析中的。 在所有的案例中， 这些案例 就是 我们 收集到的将零部件供
46、给 7 家组装工厂的 470 家供应商和承运商的案例，都是由一个第三方物流公司有组织地从几家选定的供应商中采集零部件。这些零部件要么是通过循环取货路线从交叉运转仓库运到汽车生产商（ VM），要么是是直接从供应商那运送到 VM工厂的。根据一次运载量和零部件需求总量，零部件从供应商那采集频率是一周一次或是一天多次。平均而言，一家典型的 VM 需要每天一次或多次从 50%左右的指定供应商 那 采购零部件，而并不是经常从其余供应商 那 采购。这是基于该公司的当时不断 变化的需要，采购计划间隔时间和规划的可行性而定的。如果 3 天汽车 成了 VM 的主打产品，那么这三点的每一点都会改变。 相对于 “ 物
47、理 ” 操作的相似性，这些 物流 主题中的主要不同点 就 在信息流上。在这些案例中，物流公司的采购需求信息要么是从 VM（与此同时 VM 也将信息传输给了供应商）传输过来，要么是从供应商处传输过来，或者是从物流公司传输给供应商。在第一个传输例子中，信息从 VM 传输过来可以给采购以及时的反馈信息，因为若做好的计划要 马上 变更，运送和采购计划可以在几小时内修改好。然而在最后一个传输例子中，由于以地面为通道的邮 件作为海运计划沟通方式在使用，采购内容需要 2周的修改间期。对上述方式还在使用的质疑，合理的需要（需要纸质文件，而不是电子文档）可以给出解释。即使实际的传输时间大概在 24小时到 36小时之间，但是很明显需要更快的信息流作为修改 VM 零部件需求的更快回应。另外，通常的种数繁多的包裹和硬纸板包装看来也对周转效率具有不利影响。在这些案例中，集装箱的数量从 270（ 10%的标准尺寸）到 120（ 60%的标准尺寸）不等。 总之， 所有 案例都表明要缩短总的周转周期，然而，由于在信息传递过程中的不同特点，让汽车生产商作出及时应变还是 挺困难的。应该值得注意的是这些改变通常只涉及在采购的零部件 的数 量和 种 类上，而不是真正意义上的采购主题的工艺路径上，在所有的案例中改观这 一 点都需要 很 长的时间
