Supply Chain Optimization and RFID
How will RFID impact supply chain optimization and control?

At face value, RFID represents the next generation of machine-readable labeling. It can be thought of as bar code labels on steroids. The initial difference between bar codes and RFID is the readability. RFID tags don’t require line of sight to be read, as do bar codes. There appears to be a movement toward building global data networks for reporting RFID locations and retrieving manufacturer maintained data for a tag. Such a network has been designed in concept, such as that described by the EPCGlobal organization. The next fundamental difference is that every RFID tag has a unique serial number in addition to hierarchical data identifying the original manufacturer and product information.

Being able to read the tags from any orientation promises to improve the quality of material location/movement data versus data being gathered with existing technology. This should lead to more accurate data being fed to existing ERP and data warehouse systems that drive supply optimization systems. Forecasting, master production scheduling, and distribution requirements planning systems should be able to produce better outputs based on more accurate inventory and/or material movement (shipment) data.

Another advantage RFID promises is the ability for a manufacturer to keep in contact with, or at least hear from, its material as it moves through the supply chain. When the networks, like those envisioned by EPCGlobal, become a reality, manufacturers are likely to maintain visibility of their products as they move through the supply chain. The extended visibility can be exploited several ways by supply chain optimization software.

Forecasting can be improved through the extended visibility afforded by downstream reporting of material movement. Consider the current state. Most manufacturers attempting to statistically forecast product demand measure material as it leaves their shipping docks. Either through the use of bar code technology or manually logging shipments into an ERP system, the shipment history becomes the basis for generating a statistical forecast. This type of forecast, instead of predicting product demand, actually tends to mimic the customer’s ordering pattern or more often the manufacturers order filling capability.

If the measurement point were moved to the customer’s distribution center, or stocking warehouse, the forecasting system would be working with data that was more characteristic of what is leaving a retailer’s door versus leaving the manufacturer’s door. Large scale RFID networks that could reveal a product’s location as it moves from a distribution center or warehouse to a specific store, or from a back room to the sales floor, can feed the statistical forecast with data more characteristic of the actual demand, and less characteristic of the ordering/stocking /manufacturing /shipping practices of the customer and/or manufacturer. Too often forecasts are tracking a manufacturer’s execution capability or commercial terms than the actual demand.

Shifting the material movement measuring point further down the supply chain also makes it more responsive to changes in the actual demand. When a manufacturer rewards a customer for placing large, infrequent orders, the actual consumption (by the final consumer) rate gets lost in the buffer of inventory staged along the distribution channel. If the final consumer’s demand for a product increases (or decreases) and the flow of the material is measured on the consumer’s side of the inventory buffer, the manufacturer will see the demand change before its customer’s inventory grows or shrinks enough to cause an order to be generated, changed, or just not placed. The business results of the improved forecasting include better working capital control and utilization, better customer service and reduced operating cost through manufacturing plant schedule stability.

Vendor managed inventory and consigned inventory, a reality for many suppliers, can also be managed more closely with inventory movement data being captured and transmitted through the global RFID networks in near real-time, inside the customer’s premises. Instead of waiting for a customer to notify the supplier of consignment consumption, RFID tag location can be used to record the movement from a warehouse location onto the production floor. In a VMI environment, these movements can be used to trigger restock events or inventory limit adjustments. Taking advantage of the serial number associated with each tag, the manufacturer can monitor the aging of specific pallets, cases or containers of inventory in the customer’s premises and potentially head-off product spoilage or product-aging claims.

When will all of this great data become available and be integrated with supply chain optimization systems?

That depends on many variables including:

• The technical development required to make the RFID tags cheap enough for widespread use;
• The adoption of standards for tags and data formats;
• The construction of global data networks to share the data;
• Mandates by major supply chain players (Walmart, automotive manufacturers, US Department of Defense) and so on.

What can be done today to improve supply chain optimization looking forward to RFID’s eventual widespread implementation?

Get your supply chain optimization systems in place and ready to accept the new data.

Many manufacturers are not using statistical forecasting. This relatively straightforward technology can and does improve a manufacturer’s view into the future, especially important in the most common situation where manufacturing lead times exceed their customer’s willingness to wait for product.

Use the forecasts, and their inherent uncertainty, to appropriately size inventory and safety stock levels. Mathematical models exist today that can dynamically adjust inventory limits and safety stock levels. These models use data extracted from the ERP or data warehouse systems to continuously make small adjustments to these levels as the average demand rises and falls, as the demand becomes more or less erratic, and as the financial parameters (COGS, cost of capital, etc.) change.

Use the forecast demand levels to dynamically adjust production run lengths to achieve total cost optimization. Again, well thought out models that consider cost of goods, cost of transitions, capacity constraints, interconnected production lines, and other pertinent factors can determine true economic order quantity (EOQ) for production and/or distribution. The EOQs also need to be re-evaluated constantly. As the demand changes and the cost and constraint parameters change over time, the optimum production quantities change. Setting the production run size and forgetting it, or reevaluating it quarterly or yearly leaves room for your world to change and your production plans to drift far away from optimum. Without constant monitoring and adjustment, profitability can suffer.

Use the safety stock levels, cost optimized production run lengths, and simulated inventory projections to determine what product needs to be produced or moved around the distribution system and when to do it. This is how most manufacturers utilize their production control department. If the planners and schedulers are doing this job manually, or with limited home-made tools, they can’t complete a full analysis of every SKU every day to know the most important thing to be producing or moving. They may not be considering the variability of the many factors that go into their decision making, but instead are working with averages that leave a lot of room for “emergencies” and resultant schedule changes and expediting.

These are some actions you can take today to optimize your current supply chain data; and LeadTime Technology™ can help.

LeadTime Technology™ (LTT) is a supplier of supply chain optimization software that can clean up your existing data and put it to work for your company. Demand Forecasting, Inventory Planning, Master Production Scheduling and Distribution Requirements Planning are some of the key modules LTT offers to let your data work harder for you. These modules can be implemented to get your inventory and production practices cost and service optimized. Optimized inventory levels, at the right location, go hand-in-hand with improved customer service.

So get your inventory and production/distribution activity optimized today using your current data. This will position your company to take maximum advantage of RFID data when it arrives.