Supply Chain Ch 2
Ways that the Bullwhip Effect can be alleviated?
- Collaboration (sharing info through electronic data) - Synchronizing the supply chain (coordinate planning & inventory management) - Reducing Inventory (using JIT, VMI and QR)
Forecast Error %
= ((A-F)/A) x 100 A = actual demand F = forecast demand
CPFR (Collaborative Planning, Forecasting, and Replenishment)
A process philosophy that combines the intelligence of multiple trading partners who share their plans, forecasts & delivery schedules with one another in an effort to ensure a smooth flow of goods & services across a supply chain.
Simple Linear Regression
Attempts to model the relationship between a single independent variable and a dependent variable (demand) by fitting a linear equation to the observed data. Ex. Demand might depend on how much money is spent on advertising and promotion. More money spent, higher the demand.
Multiple Linear Regression
Attempts to model the relationship between two or more independent variables and a dependent variable (demand) by fitting a linear equation to the observed data. Ex. Demand might be dependent on how much money is spent on advertising and promotion AND on the selling price charged for the product. Advertising increased & price lowered, demand will go up. Advertising increased & price increased, impact on demand is not as obvious.
Qualitative Forecasting
Based on opinion and intuition. Depends on skill and experience of forecasters & available info. Generally used when data is limited, unavailable or not relevant. Ex. New product, New Market
Personal Insight
Based on the insight of the most experienced, most knowledgeable or most senior person available. Advantage: Fastest & Cheapest forecasting technique, Can provide good forecast Disadvantage: Relies on 1 person's judgement & opinion and also their prejudices & ignorance. Unreliable- someone who is familiar with the situation often provides worse forecast than an outsider.
Forecast Error Value
FEV = A (actual demand) - F (forecast demand)
Variations in Quantitative Forecasting
-Trend Variations (movements of a variable over time. ex. technology, fashion products, toys etc.) -Random Variations (instability in data caused by random occurrences. generally short-term and caused by unpredictable events and natural disasters. ex. hurricanes cause demand for wood for repair, tree clean up, water damage etc.) -Seasonal Variation (repeating pattern of demand from year to year. ex. holiday shopping, swim suit sales etc.) -Cyclical Variation (wavelike patterns that can extend over years and cannot be easily predicted. ex. business cycles, GDP, etc. )
Qualitative Forecasting Methods
1- Personal Insight 2- Jury of Executive Opinion 3- Delphi Method 4- Sales Force Estimation 5- Customer Survey
Two basic forecasting techniques
1- Qualitative forecasting (based on opinion and intuition) 2- Quantitative forecasting (uses mathematical models and historical data)
2 important considerations about a forecast
1- Statistically speaking, the forecast will be inaccurate but it is still USEFUL. 2- The forecast is the basis for most "downstream" supply chain planning decisions, so it is critical to be as accurate as possible.
Fundamentals of Forecasting
1- Your forecast is most likely wrong 2- Simple forecast methodologies trump complex ones 3- A correct forecast does not prove your forecast method is correct 4- If you don't use the data regularly, trust it less when forecasting 5- All trends will eventually end 6- It's hard to eliminate bias, so most forecasts are biased 7- Technology is not the solution to better forecasting
Sales Force Estimation
Basically the same as the Jury of Executive Opinion except that it is performed specifically with a group of sales people. Individuals in sales bring special expertise in forecasting because they maintain closest contact with customers. Advantage: No cost to collect data, More reliable forecast because it is based on opinions of salespersons. Disadvantages: Not ideal for long-term forecasting, Salepersons may introduce bias, Salespersons may not be aware of the economic enviornment
Bad forecasting can...
Be the root cause for creating just the opposite. Garbage in = Garbage out. If the forecast is bad, everything else based on that forecast will also be bad.
Good forecasting can....
Benefit a company by facilitating more effective planning This leads to reduces inventories, reduced costs, reduced stockouts & improved customer service
Time Series Forecasting
Collect and study the past data of a given time series in order to generate probable future values for the series. Forecasts for future demand rely on understanding past demand. *PREDICTING THE FUTURE BY UNDERSTANDING THE PAST
Forecast Bias
Consistent deviation from the mean in one direction, either high or low. Consists when demand is consistently over or under forecast. SUM OF (Forecast Error) = SUM OF (Actual Demand) - SUM OF (Forecast Demand) * Negative result shows actual demand consistently less than forecast *Positive result shows actual demand was greater than forecast
Bullwhip Effect
Created when in the absence of any other information or visibility, individual supply chain participants are second-guessing what is happening with ordering patterns and potentially over-reacting.
What are forecasting & demand planning crucial components of?
Customer Satisfaction
Customer Survey
Customers are directly approached and asked to give their opinions about the particular product Advantage: It is a direct method of assessing info. Simple to administer. Does not introduce any bias or value judgement Disadvantage: Poorly formed questions could lead to unreliable info. Customers don't always answer the questionnaire. Time-consuming & costly to survey a large population
Dependent Demand
Demand for an item that is directly related to other items or finished products. Demand for these items is CALCULATED. ex. Wheels, Tires, Pedals, Seat etc.
Naive Forecasting
Demand for the next time period to be exactly the same as the demand in the last time period. Advantage: Works well for mature products & easy to determine Disadvantage: Works for ONLY mature products. Variations in any demand will create inventory issues.
Forecast
Estimate of future demand
Generally, the ........ out into the future you forecast, the ....... the deviation will likely be
Further, Greater
Linear Trend Forecasting
Imposing a best fit line across the demand data of an entire time series. Plotting the data on a chart and drawing a line through it. Advantage: Can provide accurate forecast into future even if there's no random variation Disadvantage: Seasonal & Cyclical variations are softened. This is more useful for annual forecasts than for monthly.
Negative RSFE
Indicates that the forecasts were generally too high, overestimating demand *Excess inventory & higher carrying costs are likely to occur
Positive RSFE
Indicates that the forecasts were generally too low, underestimating the demand. *Stock-outs likely to occur b/c companies cannot meet customer demand
Tracking Signal
Indicator that forecast bias is present. = RSFE / MAD *Determines if forecast is within acceptable control limits
Mean Squared Error (MSE)
Magnifies the errors by squaring each one before adding them up and dividing by the number of periods. Squaring the numbers makes them absolute. MSE = SUM OF ((A-F)^2) / n
Mean Absolute Percent Error (MAPE)
Measures the size of the error in percentage terms. Calculated as the average of the unsigned percentage error. MAPE = SUM OF (abs (A-F) / A) / n A = actual demand F = forecast demand n = number of time periods
Mean Absolute Deviation (MAD)
Measures the size of the forecast error in units. Calculated in absolute value for errors over a specified period of time. MAD = SUM OF (abs(A-F) ) / n A = actual demand F = forecast demand n = number of time periods
Goal of forecasting and demand planning process
Minimize forecast error
Exponential Smoothing
More sophisticated version of weighted moving average. Requires last period's forecast, last period's actual demand and a smoothing factor (# greater than 0 and less than 1 used as a weighting percentage). Smoothing constant must be determined based on best judgement of company experts. Advantage: Will create a forecast more responsive to trends Disadvantage: Still will lag behind trends.
Jury of Executive Opinion
People who know the most about the product and the marketplace would likely form a jury (management panel) to discuss and determine the forecast. Advantage: Decisions are enriched by experience. Companies don't have to spend time & resources collecting data by survey Disadvantage: Experts may introduce bias. They also may become biased / influenced by their colleagues
Running Sum of Forecast Errors (RSFE)
Provides a measure of forecast bias. Indicates the tendency for a forecast to be consistently higher or lower than actual demand. RSFE = SUM OF (e of t) e of t = forecast error for period t
Delphi Method
Same as the Jury of Executive Opinion except input of each of the participants collected separately so people are not influenced by one another. Done in several rounds until consensus is achieved. Advantages: Decisions are enriched by experience of experts, decisions are not a product of GROUPTHINK, very useful for new products Disadvantages: Experts may introduce bias. Companies must spend time & resources by collecting data. Can be time-consuming and is best for long-term forecasts.
What does CPFR do?
Significantly reduces the Bullwhip Effect. Provides better customer service, lower inventory costs, improved quality, reduced cycle time & better production methods.
Weighted Moving Average
Similar to simple moving average but not all historical time periods are valued equally. Proportioned amount of importance for newer months. Challenge is deciding the weight for each time period. Advantage: More accurate than simple moving average Disadvantage: This technique will still lag behind actual demand.
Forecasting
The business function that estimates future demand for products so that they can be purchased or manufactured in appropriate quantities in advance of need.
Independent Demand
The demand for an item that is unrelated to the demand for other items. Demand for these items is FORECASTED. ex. Bicycle
Forecast Error
The difference between the actual demand and the forecast demand. Can be quantified as absolute value or as a percentage.
Safety Stock
The inventory a company holds above normal needs as a buffer against delays in receipt of supply or changes in customer demand
Demand
The need for a particular product / component. Can come from various sources such as customer order, a forecast or manufacturing of another product
Reducing the Bullwhip Effect means...
The reduction of safety stocks (and associated costs) within and across the trading partners in a supply chain.
Where does the real value of CPFR come from?
The sharing of forecasts among firms, rather than firms relying on sophisticated algorithms & forecasting models to estimate demand.
Simple Moving Average
Uses calculated average historical demand during a specified number of the most recent time periods to generate the forecast. Treat all months with equal amount of importance. Advantage: Provides consistent demand over long periods of time Disadvantage: Fails to identify trends / seasonal effects. It will also lag behind actual demand.
Cause and Effect Forecasting
Uses historical relationship between an independent and dependent variable to predict future values of the dependent variable.
Quantitative Forecasting
Uses mathematical models and historical data to make forecasts. - Time Series: based on the assumption that the future is an extension of the past. HISTORICAL DATA is used to predict future demand. Most frequent among all forecasting models. - Cause and Effect: assumes that one or more factors (independent variables) predict future demand. Ex. seasonality in retail markets
Demand Planning
Where management and experts review the forecast to ensure it is aligned w/ company's strategy, business policies & knowledge and make necessary adjustments. The process of combining statistical forecasting techniques and judgment to construct demand estimates for products or services.