Intro to Supply Chain: Ch. 12

stable pattern

- a consistent horizontal stream of demands - ex. mature consumer products such as shampoo or milk

moving average

- a forecasting model that computes a forecast as the average of demands over a number of immediate past periods - always start with the most recent event

historical analogy

- a forecasting technique that uses data and experience from similar products to forecast the demand for a new product - ex. next generation electronics

shift or step change

- a one-time change in demand, usually due to some external influence on demand - ex. major product promotional campaign, sudden economic shock

intermediate term/tactical: types of decisions involved

- aggregate production plans -employee hiring and firing - planned overtime work - subcontracting - new product launches

short term/operational materials and resources: length

1-12 weeks

long term/strategic: length

1-5 years

What is the result of adaptive forecasting?

can simplify the life of the manager, but when a particular demand forecast routinely misstates actual results, it warrants some sort of management intervention

What is the primary goal in designing a forecasting process?

generate forecasts that are usable, timely, and accurate

root mean squared error (RMSE)

gives an approximation of the forecast error standard deviation

forecasting activities

integrate information gathered from the market, from internal operations, and from the larger business environment to make predictions about future demand

What does the naive model ignore?

it ignores the trend, seasonal, or other components of the historical time series, and it creates highly erratic forecasts if these components or random variations are present

What does a positive forecast bias indicate?

it indicates that over time forecasts tend to be too low

What does a negative forecast bias indicate?

it indicates that overtime forecasts tend to be too high

How is the number of periods that managers have to forecast into the future determined?

it is determined by the order-to-delivery (OTD) lead time provided by the supply chain, or the time required to source, make, and deliver the product

identify likely sources of the best data innputs

it is important to identify the potential drivers of demand and then find the data that best represents those drivers

Is it important to continually improve the forecasting process? If so, why?

it is! it is important to improve its accuracy, user-friendliness, and flexibility

In a moving average, what does increasing the number of periods (n) do?

it reduces the impact of random or atypical demands in isolated time periods, but it also reduces the sensitivity of the moving average to actual shifts in demand

How sophisticated should the demand forecasting process be?

it should be sophisticated enough to achieve acceptable levels of forecast accuracy, but simple enough so that steps involved can be understood by the users

causal studies

looks for casual relationships between leading variables and forecasted variables

What can managers observe by tracking the tracking signal over successive periods of time?

managers can observe whether undesirable trends or highly biased errors are occurring

What is the ultimate goal of demand management?

match demand and operational capacity in order to attain the business's competitive objectives

What are the forecasts and demand management plans passed on to?

materials, capacity planning, scheduling systems

With a stable demand pattern, what do the simplest time series forecasting models use?

naive model

The most sensitive simple exponential smoothing and moving average models are still...

only reactive; they do not anticipate the effects of a trend, or any seasonal or cyclical variations in demand

What are the keys to the postponable product approach?

redesign of the product and redistribution of production resources so that the products can be easily configured close to the source of demand

The longer the time period over which you have to forecast...

the greater the forecast error

What does the postponable product largely eliminate?

the need for large and complex forecasting systems, as only the demands for the relatively few individual components are forecasted, not the demands for the many different end-item configurations

tracking signal

the ratio of a running total of forecast error to MAD that indicates when the pattern of forecast error is changing significantly

What does MSE usually give a decent approximation of?

the variance of forecast errors

What are some tools that marketers have developed for the marketing research forecasting approach?

- consumer surveys - interviews - focus groups

short term/operational materials and resources: types of decisions involved

- daily production schedule - daily work schedule - purchase orders

autocorrelation

- describes the relationship of current demand with past demand - ex. if values of demand at any given time are highly correlated with demand values from the recent past, then we say that the demand is highly autocorrelated

long term/strategic: types of decisions involved

- find new sources of supply - build or sell a plant - contract for transportation services - open or close new service location

What happens if some new event changes the underlying drivers of demand in a time series analysis model?

these models will not work well

executive jdugement

- forecasting techniques that use input from high-level, experienced managers - usually better equipped to make judgements regarding long-term sales or business patterns

When do economists use historical analogy?

they use it extensively when forecasting business cycles and related developmennts

level of detail

- forecasts can be generated for an individual product, for an entire product family, or even for an entire business or industry - forecasted demand could be for a location, a county, a region, or worldwide

Delphi method

- forecasts developed by asking a panel of experts to individually and repeatedly respond to a series of questions - ex. consultants

accuracy versus cost

- greater forecast accuracy often requires greater effort and greater forecast system sophistication - it is important to weigh the costs created by forecast errors against the costs of achieving greater accuracy

Take the tactic to influence the timing or quantity of demand through pricing changes, promotions, or sales incentives: what do these moves usually intend?

- increase demand during the low periods - reduce or postpone demand during the peak periods

short term/operational materials and resources: uses of forecasts and demand management plans

- inventory planning - purchasing plans - labor scheduling

What are some of the costs of making forecasts that are too low?

- lost sales - lower product availability for customers

What four collaborative activities does the CPFR process typically consist of?

- market planning - demand and resource planning - execution - analysis

forecast accuracy

- measures how closely the forecast aligns with the observations over time - every error, whether the forecast was too high or too low, reduces accuracy

What are some of the costs of making forecasts that are too high?

- money lost in holding inventory that is never sold - lost capacity that is spent making products that no one wants - lost wages spent paying workers who are not needed

What do causal models use to predict demand?

- other independent, observed data to predict demand - concentrate on external factors that are thought to cause demand

What does the information that forecasting activities use include?

- past demand - past forecasts and their associated errors, business and economic metrics, and the judgement of experts

artificial intelligence

- refers to learning and decision making capability that stems from software algorithms - can be considered a next generation approach that combines time series analysis, causal modeling, simulation, and focused forecasting techniques - combines massive search capability, computational powder, and learning algorithms to produce more accurate demand forecasts

seasonality and cycles

- regular patterns of repeating highs and lows - seasonality may be daily, weekly, monthly, or even longer - ex. restaurants

Customer demand fluctuations cause operational inefficiencies all across the supply chain, including...

- requiring extra resources to expand and contract capacity to meet varying demand - backlogging (delivering later than originally promised) certain orders to smooth out demand fluctuations - customer dissatisfaction with the system's inability to meet all demands - buffering the system through the use of safety stocks (excess inventories) safety lead time (lead times with a cushion), or safety capacity (excess resources)

intermediate term/tactical: uses of forecasts and demand management plans

- sales and operational planning - product portfolio planning

forecasting techniques

- seek to uncover predictors of and patterns in demand and to extrapolate them to the future - suggest that some systematic forces and influencing the data - typically made up of different component drivers of demand that work together

simulation models

- sophisticated mathematical programs that offer forecasters the ability to evaluate different business scenarios that might yield different demand outcomes - this evaluation helps forecasters to better understand how different variables and drivers of demand relate to one another

long term/strategic: uses of forecasts and demand management plans

- supply chain network design - technology investments - capacity planning (investments or divestments)

What happens when the individual product forecasts are combined?

- the aggregate forecast is overall more accurate, because some of the negative errors are cancelled out by some of the positive errors - likely to reflect a more complete and unbiased picture of actual demand patterns

mean absolute deviation (MAD)

- the average size of forecast errors, irrespective of their directions - also called mean absolute error

forecast error

- the difference between a forecast and the actual demand - the "unexplained" component of demand that seems to be random in nature

time horizon

- the forecasting process should suit the period of time over which the user's current actions will affect future business performance - product demand forecast should at least match the production system's requried time

trend

- the general sloping tendency of demand, either upward or downward, in a linear or nonlinear fashion - ex. new products in the growth phase

regression analysis

- the most commonly used method for estimating relationships between leading indicators and demand - illustrates the linear relationship between the actual demand and the forecasting factors over time

forecast bias

- the tendency of a forecasting technique to continually over predict or under predict demand - also called mean forecast error

What are the four methods of statistical model-based techniques?

- time series analyses - causal studies - simulation models - artificial intelligence

postponable product

a product designed so that it can be configured to its final form quickly and inexpensively once actual customer demand is known

naive model

a simple forecasting approach that assumes that recent history is a good predictor of the near future

demand planning

the combined process of forecasting and managing customer demands to create a planned pattern of demand that meets the firm's operational and financial goals

What is a major limitation of grassroots forecasting?

- "experts" may unconsciously base their forecasts on their most recent experiences, rather than their entire set of experiences - "experts" may adjust their forecasts because of other motivations

What are the steps in designing a forecasting process?

1. identify the users and decision-making processes that the forecast will support, including time horizon, level of detail, accuracy versus cost, and fit with existing business processes 2. identify likely sources of the best data inputs 3. select forecasting techniques that will most effectively transform data into timely, accurate forecasts over the appropriate planning horizon 4. document and apply the proposed technique to the data gathered for the appropriate business process 5. monitor the performance of the forecasting process for continuous improvement

What are the three basic tactics that mangers use when trying to manage demand?

1. influence the timing or quantity of demand through pricing changes, promotions, or sales incentives 2. manage the timing of order fulfillment 3. substitute by encouraging customers to shift their orders from one product to another, or from one provider to another

What are the rules that give an indication of how situational characteristics tend to affect forecast accuracy?

1. short-term forecasts are usually more accurate than long-term forecasts 2. forecasts of aggregated demand are usually more accurate than forecasts of demand at detailed levels 3. forecasts developed using multiple information sources are usually more accurate than forecasts developed from a single source

intermediate term/tactical: length

6-18 months

weighted moving average

a forecasting model that assigns a different weight to each period's demand according to its importance

What do aggregate forecasts benefit from?

a cancellation of errors that exist in item-level forecasts

demand forecasting

a decision process in which managers predict demand patterns

marketing research

a forecasting technique that bases forecasts on the purchasing patterns and attitudes of current or potential customers

What makes a good forecasting process?

a good forecasting process acquires and analyzes information inputs in ways that address all of the relevant components of demand, while not overreacting to random changes in demand

collaborative planning, forecasting, and replenishment (CPFR)

a method by which supply chain partners periodically share forecasts, demand plans, and resource plans in order to reduce uncertainty and risk in meeting customer demand

mean squared error (MSE)

a more sensitive measure of forecast errors that approximates the error variance

exponential smoothing

a moving average approach that applies exponentially decreasing weights to each demand that occurred farther back in time

smoothing coefficient

a parameter indicating the weight given to the most recent demand

demand managament

a proactive approach in which managers attempt to influence the pattern of demand

adaptive forecasting

a technique that automatically adjusts forecast model parameters in accordance with changes in the tracking signal

grassroots forecasting

a technique that seeks inputs from people who are in close contact with customers and products

What does demand planning drive?

almost all other activities in operations management

seasonal index

an adjustment factor applied to forecasts to account for seasonal changes or cycles in demand

Take the rule that short-term forecasts are usually more accurate than long-term forecasts: why?

as the time horizon for forecasting increases, more and more potentially unknown factors can affect demand

mean percent error (MPE)

average error represented as a percentage of demand

judgement-based forecasts

built upon the estimates and opinions of people, most often experts who have related sales or operational experience

short term/operational materials and resources: demand planning units

dollar or unit sales for a given item or service at a given location

What are some unstructured forms included within big data?

e-mail and blog texts, social media, Internet click streams

Because of the squared term, the MSE gives...

exponentially more weight to larger and larger errors

When is grassroots forecasting most useful?

for developing short-term forecasts for individual products

Although both average forecast error and mean percent error are good indicators of bias, what are they not necessarily good indications of?

forecast accuracy

time series analysis models

forecasting models that compute forecasts using historical data arranged in the order of occurrence

statistical model-based forecasting techniques

forecasting techniques that transform data into forecasts using one of four methods

big data

refers to the voluminous amounts of information that are easily accessible through interconnected systems today

judgement techniques

seek to incorporate factors of demand that are difficult to capture in statistical models

What does the trend component of a time series normally result from?

some market force that causes a general rise or decline in values over time

Measures like MAD and MAPE are...

sometimes inadequate as measures of forecast accuracy

What does the forecasting process often combine?

statistical data with judgments from knowledgable sources?

materials, capacity planning, scheduling systems

systems used to manage resources and operating processes

What are some sensor data forms included within big data?

temperature, GPS, wearable technology, barcode data, videos, digital images, voice data

mean absolute percentage error (MAPE)

the MAD represented as a percentage of demand

long term/strategic: demand planning units

total dollar or unit sales for a business unit across the sales network

intermediate term/tactical: demand planning units

total dollars or unit sales for a product family in a region

What are some highly structured forms included within big data?

transaction data, location data, descriptive data

In a weighted moving average, what is more weight given to?

typically, more weight is given to more recent demand

What does a linear trend result from?

when demand rises or falls at a constant rate, describing a straight line on a graph

When is the judgement-based forecast approach useful?

when there is a lack of quantitative historical information