Big Data for Supply Chain Optimization By: Anders Richter, SAS Institute, Denmark
Agenda Demand-Driven Planning & Optimization and Big data Inventory Optimization (IO) The Matas case Results and takeaways from implementations Further readings
Demand-Driven Planning & Optimization Volume EXPLOSION OF DEMAND-RELATED DATA Bulk of this BIG Data is generated outside the company Velocity Variety
Demand-Driven Planning & Optimization THE PROCESS
Inventory Optimization TYPICAL NETWORK Supplier Store Customer Supplier DC Store Customer Store Customer Supplier Store Customer Supplier DC/ echelon lvl 2 Store/ echelon lvl 1
Inventory Optimization GOAL AND INPUT Goal with IO To find the most optimal reorder levels as to economy and which level should be ordered up to in other words finding minimum and maximum. This is done based on constrains and demand expectations on SKU level Model types SS and BS, which are minimizing the cost given the demand and constrains information Input variable Costs Demand Constrains Ordering cost, holding cost and penalty cost Expected sales in the total lead time, and the uncertainty of this expected demand Service level, service type (fill rate), batch size and minimum order quantity Combining min./max. with inventory position gives the suggested order for the SKU
Inventory Optimization INDIVIDUAL REORDER LEVEL AND ORDER UP TO LEVEL 70% 10% 20% ERP policies IO policies
Inventory Optimization INDIVIDUAL REORDER LEVEL AND ORDER UP TO LEVEL
The Matas Case About Matas 292 stores in Denmark 30,000 items Own brands + Lancôme, Clinique, etc. 2,100 employees in stores and administration
The Matas Case INCOHERENT FLOWS DC replenishment Order proposals based on DC sales Manual process correcting proposals No link to store replenishment Store replenishment Store manager controls replenishment Based on gut feeling and last 31 days of sale Very time-costly
The Matas Case THE IT SET-UP SAS Order proposal store + DC POS sales Stock levels Orders Assortment Target BI Matas DW POS sales Stock levels Store system RCM Order Assortment Order proposals store Orders SpaceMan Assortment ERP system Axapta Orders DC Suppliers
The Matas Case REPLENISHMENT NOW COHERENT FLOWS SAS forecasting (POS data) Adjust & Improve Order proposals to DC (semiautomated) Reporting on SAS quality Order proposals for stores (locked for editing)
ITEM The Matas Case CALCULATION EXAMPLE FROM MATAS Item 100059 (Eye makeup remover) Store 15288 (Greater Copenhagen) STORE SALES RECORDS, PROMOTIONS INVENTORY INFORMATION FORECASTS INVENTORY OPTIMIZATION RESULTS Forecast 42 Min 113 Std.dev. 26 Max 155 Lead time 1 Order suggestions 96 Service degree 0,99 Stock holding costs (n/a) Size of colli 12 Opening allowed N Store inventory 65
Inventory Optimization LIMITATION OF IO Limitation of IO Cannot aggregate orders on supplier level So when to use OR? When there are constrains on supplier level (minimum order amount/order size) Container optimization Push allocation Optimal distribution in case of shortages Displays
The Matas Case RESULTS AND TAKEAWAYS Total stock value reduced by 10% Out-of-stock situations reduced by 2 percentage points Able to control the out-of-stock on their ABC classification Man-hours spent on replenishment reduced by 70% Facts instead of gut feeling Coherent replenishment flows Do not forget change management Matas case study is outlined in Chapter 8
Results and Takeaways ARGUMENTS FOR STARTING WITH DDPO The system is objective It uses historical information and master data when calculating min./max. instead of being dependent on a person both with regard to gut feeling and skills Automating the creation of order proposals ensures: Time spent on generating order proposals is reduced SKUs are not forgotten, and the risk of out-of-stock situations is thus reduced Min. and max. values are always up-to-date Individual reorder level and order up to level, not one size fits all
Results and Takeaways MARKET-DRIVEN SUPPLY CHAIN BENEFITS Sense market changes 5X faster Align their supply 3X faster to fluctuations in demand With better customer service with substantially less inventory, waste and working capital (e.g., profitable supply chains) Bottom-line: Market-Driven processes are designed from the market-back -- based on sensing and shaping demand and optimizing supply
Results and Takeaways GETTING THERE Vision Phase 3 Phase two: Increase scope and automation in the process Phase one: Limited scope and creating of the data process, reap the benefits
Demand-Driven Planning & Optimization FURTHER READING Demand Sensing Demand Shaping Demand Shifting Outside-in Focused Proactive Process Collaborative Planning Lean Forecasting (FVA) Demand- Driven Market- Driven Sales & Operations Planning Supply- Driven Lean Management Supply Sensing Supply Shaping Synchronized Replenishment Inside-out Focused Reactive Process Inventory Optimization DDPO Solution overview Market Selling through the channel (pull) Supplier Selling into the channel (push)
Anders Richter Business Delivery Manager Commercial & Life Sciences Division SAS Institute Denmark E-mail: Anders.richter@sas.com Mobile: +45 27 21 28 21 Or follow me on LinkedIn