Chapter 4: Forecasting and Demand Planning

Time-series forecasting

-set of evenly spaced numerical data obtained by observing response variable at regular time period. -forecast based only on past values, no other variables important. assumes that factors influencing past and present will continue influence in future.

Exponential smoothing

-sophisticated weighted average model -needs only three numbers: Forecast for the current period t Actual demand for the current period t Weight b/w 0 and 1

Forecast error in a SC with info sharing and coordination

-supplier (10%) error -manufacturer (10%) error -wholesaler (10%) error -retailer (10%) error =dependent demand -end-use consumer =independent demand

Exponential smoothing characteristics

-the more the level changes, the larger alpha should be so that exponential smoothing can quickly adjust. -the more random the data, the smaller alpha should be so that exponential smoothing can dampen out the noise.

Causal forecasting

-tries to understand and identify the factors affecting or causing a result (for example, sales) -linear regression models -multiple regression models -econometric models -leading indicators

Short-range forecast

-up to 1 year, generally less than 3 months -purchasing, job scheduling, workforce levels, job assignments, production levels

CPFR Five-Step Process

1. Create joint objectives 2. Develop a business plan 3. Create a joint forecast 4. Agree on replenishment strategies 5. Agree on a technology partner to bring to fruition.

Time-series components

-trend -cyclical -seasonal -random

Forecasting

the art and science of predicting future events. -weather: prediction of what the whether will be. -demand: a prediction of what demand will be.

Mean Absolute Percent Error (MAPE)

the average percent error.

Selecting the smoothing constant

the objective is to obtain the most accurate forecast no matter the technique. -MAD (Mean Absolute Deviation) -MSE (Mean Squared Error) -MAPE (Mean Absolute Percent Error)

Collaborative forecasting and demand planning

two common processes: -collaborative planning, forecasting and replenishment (CPFR) -sales and operations planning (S&OP)

Qualitative forecasting methods

useful when identifying customer buying patterns, expectations, and estimating sales of new products. -jury of executive opinion -delphi method -sales force composite -market research

The Delphi Method

a consensus is developed from anonymously contributed expert info.

Jury of executive opinion

a group decision-making process, subject to bias.

Time series forecast techniques: averages

averages constructed from bigger data sets are less responsive to sudden changes. -an average that uses the eight most recent data points is less responsive than one that uses the past three.

Impact on SCM

demand forecast affects the plans made by each member of the supply chain. -independent forecasting among supply chain members causes a mismatch b/w supply and demand and gives rise to the bullwhip effect.

Random component

erratic, unsystematic, 'residual' fluctuations -due to random variation or unforeseen events -short duration and non repeating

Impact on the organization

every organizational function relies on forecasting for numerous things. -marketing: estimates of demand, future trends -finance: set budgets, predict stock prices -operations: capacity planning, scheduling, inventory levels -sourcing: make purchasing decisions, select suppliers.

Quantitative forecasting

forecasts that employ mathematical modeling to forecast demand.

Qualitative forecasting

forecasts that incorporate factors such as the decision maker's intuition, emotions, personal experiences, and value system. -advantages: can predict changes in sales patterns, can incorporate very rich sources of data. -disadvantages: large amounts of complex info, limited by availability or how recent the info, expensive and time consuming, may fail to see patterns that exist, may see patterns that don't exist.

Bullwhip effect

increasing variability of demand as one moves upstream in a supply chain. -demand variability amplified through supply chain distorted and or delayed information long lead times over batching over reactions game-playing multi-tier supplier-->manufacturer-->distributor/wholesalers-->retailers-->consumers

S&OP

is a collaborative process for generating forecasts that all functional areas agree upon.

CPFR

is a collaborative process of developing join forecasts and plans with supply chain partners.

Time series model

is a listing of data points of the variable being forecast over time. -models include: mean, moving averages, exponential smoothing, trend adjusted exponential smoothing -seasonality adjustment can also be applied.

Multiple regression

looks at the relationship b/w the independent variable and multiple dependent variables.

Sales force composite

sales force estimate based on experience, subject to bias.

Regression forecasting

set of "If, Then" statements- y=a+bx (a=y-intercept, b=slope of the line) -"if we increase promotions and/or we decrease price and/or economic indicators favor us" then sales/demand tends to increase." -provides the formula for the line that best fits through the data points.

Causal Forecasting key concept

some other variable is a leading indicator for the value you want to predict (the dependent variable value). - a good sample has many observations and potential "predictor" variables

Strategic importance of forecasting

supply-chain management-good supplier relations, advantages in product innovation, cost and speed to market. human resources-hiring, training, laying off workers capacity-capacity shortages can result in undependable delivery, loss of customers, loss of market share.

Market research

surveys and interviews used to collect preferences.

Medium-range forecast

-3 months to 3 years -sales and production planning, budgeting

Long-range forecast

-3yrs+ -new product planning, facility location, research and development

Simple moving averages

-MA is a series of arithmetic means -used if little or no trend -used often for smoothing -provided overall impression of data over time.

Forecast in traditional supply chain

-Supplier (46.4%) error -Manufacturer (33.1%) error -Wholesaler (21%) error -Retailer (10%) error =end use consumer

Factors in method selection

-amount and type of available data -degree of accuracy required -length of forecast horizon -patterns in the data

Naive approach

-assumes demand in next period is the same as demand in most recent period. -sometimes cost effective and efficient -can be good starting point.

Quantitative forecasting method strengths

-can consider many variables and complex relationships. -objective -consistent -can process large amounts of information

Qualitative forecasting method weaknesses

-cannot consider many variables -influenced by short term memory -difficulty in understanding relationships -biased (optimism, wishful thinking, political manipulation, lack of consistency)

Qualitative forecasting methods strengths

-highly responsive to latest changes in environment -can include "inside" and "soft" information difficult to quantify. -can compensate for "one-time" or unusual events -provide user with a sense of "ownership".

Demand management

-is the process of influencing demand. factors into forecasting and planning.

Quantitative forecasting method weaknesses

-only as good as the data and model. -slow to react to changing environments. -costly and time consuming to model "soft" information. -requires technical understanding.

Trend component

-persistent, overall, upward or downward pattern. -changes due to population, tech, age, culture, etc. -typically several years duration.

Seasonal component

-regular pattern of up and down fluctuations. -due to weather, customs, etc. -occurs within a single year. -multiplicative seasonal model can adjust trend data for seasonal variations in demand.

Cyclical component

-repeating up and down movements -affected by business cycle, political, and economic factors. -multiplied years duration -often causal or associative relationships.

Demand Planning Summary

1. Forecasting process choice is influenced by a variety of factors 2. Forecasts are judgement or statistical model based. 3. Both accuracy and bias should be considered 4. Demand management involves influencing customer demand. 5. Supply chains can be made more responsive to changes in customer demand.

S&OP Five-Step Process

1. Generate quantitative sales forecast 2. Marketing adjusts to forecast 3. Operations checks forecast against existing capability. 4. Marketing, operations, and finance jointly review forecast and resource issues. 5. Executives finalize forecast and capacity decisions.

Overview of quantitative approaches

1. Naive approach (time-series) 2. Moving averages (time-series) 3. Exponential smoothing (associative, causal) 4. Trend projection (associative, causal) 5. Linear regression (associative, causal)

Two groups of forecasting methods

1. Qualitative 2. Quantitative

Principles of forecasting

1. Rarely perfect 2. More accurate for groups than for individual items. 3. More accurate for shorter than longer time horizons.

Forecast process performance

1. Short term forecasts are more accurate than long term forecasts. 2. Aggregate forecasts are more accurate than detailed forecasts. 3. Information from more sources yields a more accurate forecast.

Forecast time horizons

1. Short-range forecast 2. Medium-range forecast 3. Long-range forecast

Seven basic steps of forecasting

1. determine the use of the forecast 2. select items to be forecasted 3. determine the time horizon of the forecast 4. select the forecasting model(s) 5. gather the data needed to make the forecast -common patterns include: level or horizontal, trend, seasonality, and cycles, some random variation. 6. make the forecast 7. validate and implement the results

Potential problems with moving average

1. increasing n smooths the forecast but makes it less sensitive to changes. 2. Does not forecast trends well. 3. Requires extensive historical data.

Forecasting vs planning

1. the process of predicting events 2. the process of selecting actions in anticipation of the forecast. -involves the following: scheduling existing resources. determining future resource needs acquiring new resources

Mean Absolute Deviation (MAD)

How much the forecast missed the target.

Mean Squared Error (MSE)

The square of how much the forecast missed the target.