Guide to Supply Chain Management

by David Jacoby

  A UK call-centre operator uses APS to manage streams of customer calls, e-mails, web chat and web support services from around the world, and it personalises its response to every enquiry. Operations like this need APS to determine appropriate staffing levels and shift structure and work assignments, given the peaks and valleys of traffic and the varying profit margin of the different types of orders that come in at different times of the day.

  Materials requirements planning and production scheduling

  Materials requirements planning (MRP) determines the right amount of material needed when demand is triggered by an item that requires a component. MRP explodes the original demand into the items contained in its bill of materials, and then orders each component or part within the lead time needed so that all the parts arrive in time for assembly. This leads to the right amount of production and therefore fewer inventory imbalances. Since MRP became popular in the 1970s, its penetration has deepened, and today even most small manufacturers have implemented MRP systems.

  Warehouse management systems

  Warehouse management systems (WMS) help make pick-pack operations more efficient and reduce the required footprint of the warehouse. They determine the routing that minimises the resources required to pick (or put away) material in a warehouse. To do that, they optimise the storage plan of the material based on its velocity. Like TMS, storage plans can be either static or dynamic. In large warehouses, WMS work in conjunction with automated guided vehicles (AGVs), conveyors and other equipment such as automated storage and retrieval systems (ASRS) in order to co-ordinate materials’ flow in and out of the shelves using the least amount of resources. Large auto part warehouses, such as those for Ford or GM, and high-volume third-party warehouse companies also use their WMS to plan waves of picking.

  Transportation management systems

  Transportation optimisation solutions fall into seven broad categories, and sometimes interface with wireless communication technologies, optical character recognition systems and resource planning systems:

  Carrier management systems help shippers that do not own their own fleets to interface more efficiently with their transportation providers. These systems can provide rating, tendering, payment and freight bill auditing capabilities. The abbreviation TMS is often used to refer to this type of carrier transportation management system.

  Routing and scheduling systems offer features such as territory planning, load consolidation planning and execution, dynamic and static vehicle and driver assignment (the difference is whether or not the routing solution is limited to a fixed assignment of certain drivers to certain trucks or certain trucks to certain routes, or whether it can find the optimal solution without these constraints), routing, scheduling around time windows, mapping and in some cases what-if modelling capabilities. Some vendors of these offer planning systems that are used only when the network configuration changes, while others offer execution systems that are meant to be run every day or every time the route is run. Some systems offer both territory or area planning and dynamic route generation capabilities.

  Real-time dispatching systems include some capability to balance loads, either inbound and outbound, or from vehicle to vehicle. Some have the capability to edit routes manually once they are created, while others do not. Some high-end systems offer vehicle tracking after the route has begun, while others only generate the route itself.

  Tracking and communication systems involve a wireless hardware device, which can be on the vehicle or on the driver, as in a handheld device or PDA (personal digital assistant such as a mobile phone or BlackBerry). The cost of the hardware can be balanced against the benefits of improved service and the ability to divert and change orders once the vehicle is in transit.

  Order verification and reporting systems allow central dispatchers or customer service managers to see the real-time status of orders as they are delivered.

  Fleet management systems monitor equipment as a whole, or components of the equipment, such as engine performance, fuel consumption and compliance with local requirements to pay fuel and mileage taxes based on places driven to.

  Trade compliance systems, applications that help to determine the duties and documentation required when shipping internationally, can also provide real-time information about the economics of end-to-end supply chains. Trade compliance systems increasingly provide landed costs and the ability to view breakdowns of shipment costs showing fuel costs, labour rates, real-estate and facility costs, cross-dock economics, customs clearance preferences and handling import/export paperwork. Some of these tools allow shippers to recalculate the cost of end-to-end supply chains by changing parameters such as the port or shipment size.

  The right choice of a TMS depends on factors such as service requirements, fleet size (if there is proprietary equipment), time spent on the road versus at drop-off locations, and the flexibility of the workforce to implement change.

  Wireless technologies such as RFID and optical character recognition interface with TMS systems to deliver real-time positioning information that enhances transportation security and efficiency.

  Wireless communication technologies

  Technologies such as radio frequency identification (RFID), a microchip-based signalling system whereby readers are used to identify items that are tagged, are increasing the ability of companies to track inventory flowing through the pipeline, and to personalise their relationship at the moment of delivery with their customers. If and when it is used at the item level on a large scale, shoppers may trigger the replenishment of grocery store items when they pick up the item from the shelves, and they could be recognised when they come through the door by an RFID chip in their loyalty card. Airline passengers could walk nonchalantly through the airport if an RFID chip designated them as low-risk passengers to automated inspectors. Many airlines and airports, from Amsterdam to Tokyo to Las Vegas, are testing RFID for baggage handling. Applications will focus on high-value transactions (such as expedited cargo) and convenience.

  In retail supply chains, RFID has had some success with companies such as Wal-Mart, Tesco and Marks and Spencer, and has in some cases displaced the electronic article surveillance tag. RFID has also started to replace the barcode in libraries and document applications. Nederlandse Bibliothek Dienst, the Netherlands library service, piloted RFID at the public library in Eindhoven in 2002. The Vatican has tagged 30,000 items. Other libraries using this system include the Catholic University of Leuven in Belgium, Olin College in Boston, MA, and the University of Nevada at Las Vegas.

  Optical character recognition

  Optical character recognition (OCR) is being used to track cargo and improve operational efficiency.4 China’s Dalian Dayaowan Port used to experience lengthy loading and uploading and could process only 40 containers per hour. Moreover, its data input error rate was high. It implemented an OCR reader to read the licence plates on the trucks entering and departing. The system increased throughput to 180 containers per hour, lowered operating cost and achieved 95% data accuracy.

  Distribution resource planning

  Distribution resource planning (DRP) systems plan the amount of inventory that is needed in multi-tier distribution networks, for example where there are two warehouses between the source and the customers. Similar to MRP, DRP uses the demand (in this case, a forecast) to calculate how much product needs to be resupplied and in what timeframe. Then, based on that information, it places orders on the warehouses. DRP systems are often viewed as a cause of the bullwhip effect for two reasons: first, in a multi-tier warehouse network, one of the warehouses has larger order quantities than the others, so one small order at the point of consumption creates larger orders at the farther warehouse; and second, DRP is based on a forecasting approach, as opposed to efficient consumer response (ECR) principles.

  Customer relationship management

  CRM systems gather and manage data on customer interactions in order to increase the quality of future interactions.
At a minimum, CRM is just a mechanism for gathering and maintaining customer transaction data, especially for web transactions since they are increasingly common. However, a robust CRM system should help realise each supply chain strategy. During rationalisation, CRM should provide a comparison of the customer profile, or needs, with the product purchased, to facilitate a gap analysis that sheds light on which features of the product are most useful. In the synchronisation phase, CRM should connect with back-end order fulfilment applications to allow flawless and timely execution. In the customisation phase, it should allow for customers to include personal information such as their credit card details, addresses and notes in their order, and for this to be remembered the next time if they wish. Finally, in the innovation phase, CRM should provide a creative user interface and a base of data that can be mined to identify new and creative product offerings.

  Amazon has extraordinary CRM capabilities in the four ways mentioned above. It creates a detailed transaction database. It tracks what people buy compared with what people like them buy, thereby allowing a value analysis. The value analysis is presented directly to the consumer, who can decide how to best spend his or her money, rather than to Amazon personnel. The system connects with back-end order fulfilment operations; it even offers different shipping times and announces the cut-offs by which the shipping dates will change, which allows customers to decide when to make their purchase. Lastly, it offers a wide array of continuously changing choices to the customer.

  Cost, pricing and risk applications

  SCM needs five cost and pricing capabilities that extend beyond the normal capabilities of the classic enterprise resource planning (ERP) financial and accounting modules. In the rationalisation stage, companies need to know the total cost of ownership (TCO) in order to reduce lifetime cost. Heavy equipment that is routinely overhauled is one example of lifetime cost.

  FedEx calculated the cost per cycle (take-off and landing) and the cost per operating hour of its jet engines in order to reduce total lifetime cost. Since each overhaul basically restores the engines to brand-new condition, it is tricky to separate a cause of failure between the original equipment manufacturer (OEM), the overhaul vendor and the severity of the flight mission (for example, repeated take-offs and landings are more stressful on the engines than a long cruise at high altitudes). FedEx developed maintenance and failure analysis models, and analysed the cost by type of overhaul and repair vendor to ascertain the normalised cost per cycle and cost per hour.

  At the synchronisation stage, companies need to measure and manage risk to avoid supply chain disruption. Supplier risk, economic risk, input cost risk and others need to be reported to shareholders as part of the Sarbanes-Oxley legislation. A market intelligence system can help mitigate the impact of imbalances between supply and demand by providing visibility to shortages and price abnormalities.

  At the customisation stage, companies should use activity-based costing (ABC) and practise yield pricing. ABC helps to allocate overhead to the prime consumer, thus helping to compute customer profitability. Yield pricing helps to simultaneously satisfy multiple customer segments.

  At the innovation stage, systems are needed to support revenue management, that is, deciding how to price and prioritise orders or shipments based on their individual profitability. Priceline, an online airline ticket reseller, has a web interface that allows customers to place bids and each offer is evaluated on its own merits by the airlines.

  E-commerce

  At the basic level, e-commerce requires order fulfilment through an unseen warehouse. Such virtual fulfilment is based on HTML and XML protocols interfaced with a WMS. An important improvement is the proper functioning of available-to-promise features, a hallmark of the synchronisation stage. Amazon and most other e-tailers offer specific promise dates for shipping and delivery, thanks to back-end inventory availability data and interfaces with logistics providers’ cut-off and delivery times. Additional customisation features include, for instance, Amazon’s referral features (“readers who bought this book also bought this other book …”), which can increase sales considerably. The accumulated mass of customisation data can be used to rapidly develop and introduce new products (innovation).

  EDI/XML

  Electronic data interchange (EDI) has been the standard for e-commerce and other transaction set transfers between trading partners for decades. The American National Standards Institute (ANSI) and UN/EDIFACT protocols have specified the protocols for the transfer of everything from requests for quotations (ANSI 840) to advance shipping notices (ANSI 856). XML is replacing these protocols as web-based transactions dominate offline transactions. For example, when the European company mentioned on page 181 embarked on the second round of its e-procurement transaction efficiency programme, it chose UNSPSC item classification codes and XML messaging and ordering protocols. The number of SKUs covered rose from 40,000 to 170,000, costs decreased and ordering fill rates increased.

  Product life-cycle management

  As mentioned in Chapter 9, a key function of PLM is to process and facilitate easy storage and retrieval of changes to the product design and specifications. At the rationalisation stage, PLM systems need to be able to help identify and reduce product complexity to facilitate cost reduction through simplification. At the synchronisation stage, the PLM output should be used in the S&OP process to plan for emerging (new) and declining (old) items and services. In the customisation stage, PLM should be used to plan value-added after-sales products and services related to the item(s). And in the innovation stage, PLM should be used to facilitate rapid and repeated product design changes.

  Cochlear, a global hearing solutions provider, implemented a PLM solution from Parametric Technology Corporation to accelerate its design review process and streamline the process of administering engineering change orders. The company considers the programme to have had a significant impact on its ability to share documents globally and therefore enhance its speed to market with new products. By replacing manual processes with automated ones, the company works more collaboratively and concurrently than in the past, when paper documents needed to be assembled and passed from one department to another. It relies on its PLM solution to help it manage the stage-gate product development process more tightly, transmit complete sets of documentation to manufacturing for pilot programmes and reduce the potential for errors throughout the entire process.

  Integration of the components

  Effective supply chain support relies foremost on full functionality and integration between core ERP modules such as finance and engineering. Several particularly important transactional modules are as follows:

  Purchasing and accounts payable need to be complete and up to date to permit supplier spend consolidation in the rationalisation stage.

  The quality of order management and APS data needs to be high to allow collaborative inventory planning in the synchronisation stage.

  Finance and in particular cost accounting (cost of service) need to be fully functioning to allow ABC to support customer profitability analysis in the customisation stage.

  In addition, best-of-breed applications (those that are not a part of the core ERP system) should be integrated with each other to allow the full success of critical initiatives at each stage. These typically include marketing, sales, CRM, SCM, SRM, WMS and engineering components.

  Marketing applications such as yield pricing need to link with production planning and scheduling to permit the rapid introduction of new products. Aramex, a Middle Eastern transportation and logistics company, can introduce new products particularly quickly because of linkage of scheduling and pricing databases.

  Sales applications need to link with distribution information systems to allow customer-specific handling in the customisation stage.

  CRM, SCM and SRM need to be tied together to allow end-to-end supply chains to become a reality. A leading oil company helps its suppliers process its customers’ orders by operating bo
th its call centre and its accounts payable via the same ERP system.

  WMS picking status, which indicates what stage of processing orders are in as they move from picking through packing and shipping, must be integrated with TMS and cross-docking applications to provide timely inventory availability information. Georgia Pacific designed a custom application to link its WMS to order management in its ERP system, enabling it to transform its warehouse into a cross-dock and thereby reduce inventory.

  Engineering, APS, purchasing and distribution modules all need to be working properly to permit concurrent engineering. David Hastings, CFO of US company Incyte Pharmaceuticals, explains that “siloed organisation can kill productivity”. Chemists and biologists are often in different parts of the country, whereas at Incyte they are down the hall from each other. Cross-enterprise collaboration accelerates new product development. Incyte and Pfizer collaborated to improve the research process. This improved IT interfaces and enhanced communication between the two companies to make a seamless partnership.

  12 Measuring success

  “You get what you measure,” goes the adage. This chapter is about which metrics are good barometers of supply chain management effectiveness, and what target levels make sense.

  Until the mid-1960s, metrics for measuring supply chain processes were uncommon. In 1966, only 58% of companies surveyed even measured inventory turnover or obsolete stock.1 The cost of overproduction was rarely counted in companies’ books, so the velocity of material as it flowed through the supply chain did not matter.

  The growing awareness of supply chain concepts, combined with ambiguity over the boundaries of SCM, led to a proliferation of low-level process metrics whose linkage to value creation was tenuous. Initially, supply chain metrics focused on inventory levels. Then, when life-cycle cost became a popular concept, the focus shifted to total supply chain cost. When the theory of constraints became popular, metrics such as throughput and inventory dollar-days gained currency. When lean accounting became popular, metrics focused on activity-based costs (ABC).2 The “plan-source-make-deliver” model introduced by the Supply Chain Council brought hundreds of metrics. The cumulative pile of metrics was so extensive that many people began to categorise them, in many cases by the old (pre-supply chain) functional boundaries.