Forecasting
Fortune Council survey, Nov 2005 Forecasting Survey How far into the future do you typically project when trying to forecast the health of your industry? less than 4 months 3% 4-6 months 12% 7-12 months 28% > 12 months 57% Fortune Council survey, Nov 2005
Indices to forecast health of industry Consumer price index 51% Consumer Confidence index 44% Durable goods orders 20% Gross Domestic Product 35% Manufacturing and trade inventories and sales 27% Price of oil/barrel 34% Strength of US $ 46% Unemployment rate 53% Interest rates/fed funds 59% Fortune Council survey, Nov 2005
Forecasting Importance Improving customer demand forecasting and sharing the information downstream will allow more efficient scheduling and inventory management Boeing, 1987: $2.6 billion write down due to “raw material shortages, internal and supplier parts shortages” Wall Street Journal, Oct 23, 1987
Forecasting Importance “Second Quarter sales at US Surgical Corporation decline 25%, resulting in a $22 mil loss…attributed to larger than anticipated inventories on shelves of hospitals.” US Surgical Quarterly, Jul 1993 “IBM sells out new Aetna PC; shortage may cost millions in potential revenue.” Wall Street Journal, Oct 7, 1994
Principles of Forecasting Forecasts are usually wrong every forecast should include an estimate of error Forecasts are more accurate for families or groups Forecasts are more accurate for nearer periods.
Important Factors to Improve Forecasting Record Data in the same terms as needed in the forecast – production data for production forecasts; time periods Record circumstances related to the data Record the demand separately for different customer groups
Forecast Techniques Extrinsic Techniques – projections based on indicators that relate to products – examples Intrinsic – historical data used to forecast (most common)
Forecasting Forecasting errors can increase the total cost of ownership for a product - inventory carrying costs - obsolete inventory - lack of sufficient inventory - quality of products due to accepting marginal products to prevent stockout
Forecasting Essential for smooth operations of business organizations Estimates of the occurrence, timing, or magnitude of uncertain future events Costs of forecasting: excess labor; excess materials; expediting costs; lost revenues
Forecasting Predicting future events Usually demand behavior over a time frame Qualitative methods Based on subjective methods Quantitative methods Based on mathematical formulas
Strategic Role of Forecasting Focus on supply chain management Short term role of product demand Long term role of new products, processes, and technologies Focus on Total Quality Management Satisfy customer demand Uninterrupted product flow with no defective items Necessary for strategic planning
Strategic Role of Forecasting Focus on supply chain management Short term role of product demand Long term role of new products, processes, and technologies Focus on Total Quality Management Satisfy customer demand Uninterrupted product flow with no defective items Necessary for strategic planning
Components of Forecasting Demand Time Frame Short-range, medium-range, long-range Demand Behavior Trends, cycles, seasonal patterns, random
Time Frame Short-range to medium-range Long-range Daily, weekly monthly forecasts of sales data Up to 2 years into the future Long-range Strategic planning of goals, products, markets Planning beyond 2 years into the future
Demand Behavior Trend Cycle Seasonal pattern gradual, long-term up or down movement Cycle up & down movement repeating over long time frame Seasonal pattern periodic oscillation in demand which repeats Random movements follow no pattern
Forms of Forecast Movement Time (a) Trend (d) Trend with seasonal pattern (c) Seasonal pattern (b) Cycle Demand Random movement Figure 8.1
Forecasting Methods Time series Qualitative methods Delphi method Regression or causal modeling Qualitative methods Management judgment, expertise, opinion Use management, marketing, purchasing, engineering Delphi method Solicit forecasts from experts
Time Series Methods Statistical methods using historical data Moving average Exponential smoothing Linear trend line Assume patterns will repeat Naive forecasts Forecast = data from last period
Moving Average Average several periods of data Dampen, smooth out changes Use when demand is stable with no trend or seasonal pattern stock market analysis - trend analysis
Moving Average Average several periods of data Dampen, smooth out changes Use when demand is stable with no trend or seasonal pattern Sum of Demand In n Periods n
Simple Moving Average ORDERS MONTH PER MONTH Jan 120 Feb 90 Mar 100 Apr 75 May 110 June 50 July 75 Aug 130 Sept 110 Oct 90 ORDERS MONTH PER MONTH
Simple Moving Average Daug+Dsep+Doct MAnov = 3 90 + 110 + 130 = 3 Jan 120 Feb 90 Mar 100 Apr 75 May 110 June 50 July 75 Aug 130 Sept 110 Oct 90 ORDERS MONTH PER MONTH 3 Daug+Dsep+Doct MAnov = = 90 + 110 + 130 3 = 110 orders for Nov Example 8.1
Simple Moving Average ORDERS THREE-MONTH Jan 120 – Feb 90 – Mar 100 – Apr 75 103.3 May 110 88.3 June 50 95.0 July 75 78.3 Aug 130 78.3 Sept 110 85.0 Oct 90 105.0 Nov – 110.0 ORDERS THREE-MONTH MONTH PER MONTH MOVING AVERAGE Example 8.1
Simple Moving Average = 91 orders for Nov Jan 120 – Feb 90 – Mar 100 – Apr 75 103.3 May 110 88.3 June 50 95.0 July 75 78.3 Aug 130 78.3 Sept 110 85.0 Oct 90 105.0 Nov – 110.0 ORDERS THREE-MONTH MONTH PER MONTH MOVING AVERAGE = 90 + 110 + 130 + 75 + 50 5 = 91 orders for Nov Example 8.1
Simple Moving Average ORDERS THREE-MONTH FIVE-MONTH Jan 120 – – Feb 90 – – Mar 100 – – Apr 75 103.3 – May 110 88.3 – June 50 95.0 99.0 July 75 78.3 85.0 Aug 130 78.3 82.0 Sept 110 85.0 88.0 Oct 90 105.0 95.0 Nov – 110.0 91.0 ORDERS THREE-MONTH FIVE-MONTH MONTH PER MONTH MOVING AVERAGE MOVING AVERAGE Example 8.1
Weighted Moving Average Adjusts moving average method to more closely reflect data fluctuations
Weighted Moving Average WMAn = i = 1 Wi Di where Wi = the weight for period i, between 0 and 100 percent Wi = 1.00 Adjusts moving average method to more closely reflect data fluctuations
Weighted Moving Average Example MONTH WEIGHT DATA August 17% 130 September 33% 110 October 50% 90
Weighted Moving Average Example MONTH WEIGHT DATA August 17% 130 September 33% 110 October 50% 90 November forecast WMA3 = 3 i = 1 Wi Di = (0.50)(90) + (0.33)(110) + (0.17)(130) = 103.4 orders 3 Month = 110 5 month = 91
Exponential Smoothing Averaging method Weights most recent data more strongly Reacts more to recent changes Widely used, accurate method
Exponential Smoothing Ft +1 = Dt + (1 - )Ft where Ft +1 = forecast for next period Dt = actual demand for present period Ft = previously determined forecast for present period = weighting factor, smoothing constant Averaging method Weights most recent data more strongly Reacts more to recent changes Widely used, accurate method
Forecast for Next Period Forecast = (weighting factor)x(actual demand for period)+(1-weighting factor)x(previously determined forecast for present period) 0 > <= 1 Lesser reaction to recent demand Greater reaction to recent demand
Seasonal Adjustments Repetitive increase/ decrease in demand Use seasonal factor to adjust forecast
Seasonal Adjustments Repetitive increase/ decrease in demand Use seasonal factor to adjust forecast Seasonal factor = Si = Di D = demand for period/sum of demand
Seasonal Adjustment DEMAND (1000’S PER QUARTER) YEAR 1 2 3 4 Total 2008 12.6 8.6 6.3 17.5 45.0 2009 14.1 10.3 7.5 18.2 50.1 2010 15.3 10.6 8.1 19.6 53.6 Total 42.0 29.5 21.9 55.3 148.7 DEMAND (1000’S PER QUARTER) YEAR 1 2 3 4 Total
Seasonal Adjustment S1 = = = 0.28 D1 D 42.0 148.7 S2 = = = 0.20 D2 2008 12.6 8.6 6.3 17.5 45.0 2009 14.1 10.3 7.5 18.2 50.1 2010 15.3 10.6 8.1 19.6 53.6 Total 42.0 29.5 21.9 55.3 148.7 DEMAND (1000’S PER QUARTER) YEAR 1 2 3 4 Total S1 = = = 0.28 D1 D 42.0 148.7 S2 = = = 0.20 D2 29.5 S4 = = = 0.37 D4 55.3 S3 = = = 0.15 D3 21.9
Seasonal Adjustment DEMAND (1000’S PER QUARTER) YEAR 1 2 3 4 Total 2008 12.6 8.6 6.3 17.5 45.0 2009 14.1 10.3 7.5 18.2 50.1 2010 15.3 10.6 8.1 19.6 53.6 Total 42.0 29.5 21.9 55.3 148.7 DEMAND (1000’S PER QUARTER) YEAR 1 2 3 4 Total Si 0.28 0.20 0.15 0.37
Seasonal Adjustment Forecast for 1st qtr 2011 2008 12.6 8.6 6.3 17.5 45.0 2009 14.1 10.3 7.5 18.2 50.1 2010 15.3 10.6 8.1 19.6 53.6 Total 42.0 29.5 21.9 55.3 148.7 DEMAND (1000’S PER QUARTER) YEAR 1 2 3 4 Total Si 0.28 0.20 0.15 0.37 Forecast for 1st qtr 2011 Forecast for 2011 using simple 3 year moving ave
Forecast Accuracy Find a method which minimizes error Error = Actual - Forecast
Forecast Control Reasons for out-of-control forecasts Change in trend Appearance of cycle Weather changes Promotions Competition Politics
Supply Chain Management
Supply Chain Management First appearance – Financial Times Importance - → Inventory ~ 14% of GDP → GDP ~ $12 trillion → Warehousing/Trans ~ 9% of GDP → Rule of Thumb - $12 increase in sales to = $1 savings in Supply Chain 1982 Peter Drucker – last frontier Supply Chain problems can cause ≤ 11% drop in stock price Customer perception of company
SCOR Reference: www.supply-chain.org
End-to-End Supply Chain Supplier Customer Suppliers’ Supplier Source Internal or External Your Company Return Deliver Make Plan Customers’ Customer SCOR reference model Because SCOR is does not represent organizations, but rather activities, it is ‘boundaryless’. This is intentional. When you are considering how a supply-chain operates, you don’t want to ‘stop’ describing it when you must look at the activities of some participants. You can, in effect, look all the way from a grain of sand to a finished computer. In looking at various ways to describe this, you can use an anecdote that SCOR looks from “Cow to Cone” for Ice Cream, or if you feel particularly humorous, “from Stump to Rump” for toilet paper. Different trainers use different anecdotes, but you need to get across to the audience that by using Standard names Standard interconnects Boundaries only defined by process start/stop points You have the ability to truly optimize the performance of very, very large systems, cutting waste, cycle time, and improving cash consumption. This is a second unique feature of the SCOR model. Whether from Cow to Cone or from Rock to Ring SCOR is not limited by organizational boundaries Copyright © Supply Chain Council, 2008. All rights reserved 45
End-to-End Supply Chain Customer’s Customer Customer MP3 Company Supplier Supplier’s Supplier Sub assemblies Manufacturer Retailer Consumer Components Source Deliver Make Process, arrow indicates material flow direction Source, Make, Deliver connections from suppliers to their customers. Note that the Retailer does not have a Make process as it is simply sourcing goods and delivering those to their customers. (If asked: yes, retail can do Make, for example preparing food, repackaging in the deli or meats departments or cutting wood in the lumber department). Copyright © Supply Chain Council, 2008. All rights reserved 46
Supply Chain “The global network used to deliver products and services from raw materials to end customers through an engineered flow of information, physical distribution, and cash.” APICS Dictionary
Supply Chain Uncertainty Forecasting, lead times, batch ordering, price fluctuations, and inflated orders contribute to variability Inventory is a form of insurance Distorted information is one of the main causes of uncertainty Bullwhip effect
Information in the Supply Chain Centralized coordination of information flows Integration of transportation, distribution, ordering, and production Direct access to domestic and global transportation and distribution channels Locating and tracking the movement of every item in the supply chain - RFID
Information in the Supply Chain Consolidation of purchasing from all suppliers Intercompany and intracompany information access Data acquisition at the point of origin and point of sale Instantaneous updating of inventory levels Visibility
Electronic Data Interchange Computer-to-computer exchange of business documents in a standard format Quick access, better customer service, less paperwork, better communication, increased productivity, improved tracing and expediting, improves billing and cost efficiency
Bar Codes Computer readable codes attached to items flowing through the supply chain Generates point-of-sale data which is useful for determining sales trends, ordering, production scheduling, and deliver plans 1234 5678
IT Issues Increased benefits and sophistication come with increased costs Efficient web sites do not necessarily mean the rest of the supply chain will be as efficient Security problems are very real – camera phones, cell phones, thumb drives Collaboration and trust are important elements that may be new to business relationships
Suppliers Purchased materials account for about half of manufacturing costs Materials, parts, and service must be delivered on time, of high quality, and low cost Suppliers should be integrated into their customers’ supply chains Partnerships should be established On-demand delivery (JIT) is a frequent requirement - what is JIT and does it work?
Sourcing How does single source differ from sole source? Relationship between customers and suppliers focuses on collaboration and cooperation Outsourcing has become a long-term strategic decision Organizations focus on core competencies Single-sourcing is increasingly a part of supplier relations How does single source differ from sole source?
Distribution The actual movement of products and materials between locations Handling of materials and products at receiving docks, storing products, packaging, and shipping Often called logistics Driving force today is speed
Distribution Centers and Warehousing DCs are some of the largest business facilities in the United States Trend is for more frequent orders in smaller quantities Flow-through facilities and automated material handling Final assembly and product configuration (postponement) may be done at the DC
Warehouse Management Systems Highly automated systems A good system will control item slotting, pick lists, packing, and shipping Most WMS include transportation management (load management/configuration), order management, yard management, labor management, warehouse optimization
Vendor-Managed Inventory Not a new concept – same process used by bread deliveries to stores for decades Reduces need for warehousing Increased speed, reduced errors, and improved service Onus is on the supplier to keep the shelves full or assembly lines running variation of JIT Proctor&Gamble - Wal-Mart Home Depot
Collaborative Distribution and Outsourcing Collaborative planning, forecasting, and replenishment (CPFR) Allows suppliers to know what is really needed and when Electronic-based exchange of data and information Europe’s ECR/QR
Transportation Common methods are railroads, trucking, water, air, intermodal, package carriers, and pipelines
Railroads 150,000 miles in US Low cost, high-volume Improving flexibility intermodal service double stacking Complaints: slow, inflexible, large loads Advantages: large/bulky loads, intermodal
Trucking Most used mode in US -75% of total freight (volume not total weight) Flexible, small loads Consolidation, Internet load match sites Single sourcing reduces number of trucking firms serving a company Truck load (TL) vs. Less Than Truck Load (LTL)
Air Rapidly growing segment of transportation industry Lightweight, small items Quick, reliable, expensive (relatively expensive depending on costs of not getting item there) Major airlines and US Postal Service, UPS, FedEx, DHL
Package Carriers UPS, US Postal Service, FedEx Ground Significant growth driven by e-businesses and the move to smaller shipments and consumer desire to have it NOW Use several modes of transportation Innovative use of technologies in some cases Online tracking – some better than others
Intermodal Combination of several modes of transportation Most common are truck/rail/truck and truck/water/rail/truck Enabled by the use of containers – the development of the 20 and 40 foot containers significantly changed the face of shipping
Early form of intermodal transport and cross docking Switching Milk Cans from a Farmer’s Buggy to a Truck on a Rural Road in North Carolina, 1929 Early form of intermodal transport and cross docking
Water One of oldest means of transport Low-cost, high-volume, slow (relative) Security - sheer volume - millions of containers annually Bulky, heavy and/or large items Standardized shipping containers improve service The most common form of international shipping
Pipelines Primarily for oil & refined oil products Slurry lines carry coal or kaolin High initial capital investment Low operating costs Can cross difficult terrain
Global Supply Chain Free trade & global opportunities Nations form trading groups No tariffs or duties Freely transport goods across borders Security!!