Untitled Flashcards Set

• Define forecasting, give 3 examples:

  • The scientific process of predicting a future event

  • Ex: weather forecast, sports games, sales forecast, traffic, etc

  • We study demand/sales forecasting

• Give 5 differences between short-range and long-range forecasting:

  • Short-range: 

    • Predict a few days into the future

    • More accurate

    • Used to make day-to-day operational decisions

    • Uses quantitative techniques

    • Used in the maturity & decline phase of PLC

  • Long-range: 

    • Predict a few years into the future

    • Less accurate

    • Used to make strategic long-term decisions

    • Uses qualitative techniques

    • Used in the Intro & Growth phase of PLC

• What is PLC? How is it related to forecasting? (Draw the diagram)

  • Product life cycle shows relationship between sales/demand and time. 

  • Introduction and growth require longer forecasts than maturity and decline.

  • Introduction - Growth - Maturity - Decline

  • Introduction it slightly grows (no one knows about it), growth it grows rapidly(people start liking it/learning about it), maturity/stabilization it flattens out (competitors), and decline it declines 

  • Short-range forecasting is used during maturity and decline, while long-range forecasting is used during the intro and growth phases. 

  • 
    

• Why is forecasting important? Give at least 4 reasons: (at least 1 sentence in your own words about each)

  • Reliable forecasting helps in better

    • Supply chain Management: lets us know how much we need to order (supplies, materials, etc) and when

    • Capacity planning: lets us know how much capacity we have to store, produce, sell, etc

    • Human resource/personnel planning: lets us know how many employees we need and/or where they should be

    • Scheduling: lets us know when we should schedule our employees, on what days, and for how long

    • Market positioning: lets us know when we should release products, advertisements, etc according to market conditions

• List 4 qualitative forecasting techniques:

  • Jury of Executive Opinion (group of CEOs, CFOs, and such)

  • Delphi method

  • Sales force composite

  • Market Survey (questionnaires by email or phone, fill it out)

• Write short notes on the following:

  • Jury of executive opinion:

    • Involves a small group of high-level experts and managers

    • Relies on managerial experience and expertise (intuition and gut feeling)

    • Relatively quick

    • ‘Group think’ disadvantage

  • Delphi method:

    • The iterative group process continues until consensus is reached

    • Three types of participants: Decision makers, staff, and respondents

  • Sales Force Composite:

    • A method of forecasting future demand for a product by adding together what each member of the sales force expects to be able to sell in his/her territory

    • Each salesperson projects his/her sales

    • Combined at district and national levels

    • Sales reps know customer’s wants

    • May be overly optimistic

    • Ex: Lexus dealers

• What is a time series? Describe the components and draw a pic

  • A time series is a series of interspersed data collected at regular time intervals. For example, the stock price of a company plotted on the Y axis and the time on the x axis. The four components of time series are

    • Trend: gradual upward or downward overall movement of the graph

    • Seasonality: regular ups and downs of the graph depending on the seasons (weekdays vs weekends, summer vs winter)

    • Cyclicals: other cycles related to macroeconomic events, (inflation, economic depressions, currency fluctuations, etc), weather incidents, global events, etc

    • Random errors: unsystematic, unpredictable small errors or fluctuations

• Why is location a strategically important decision?

  • One of the most important decisions a firm makes

  • Increasingly global in nature

  • Significant impact on fixed and variable costs

  • Decisions made relatively infrequently 

  • Long-term decisions

  • Once committed to a location, many resource and cost issues are difficult to change

• List the seven factors that affect location decisions

  • Labor productivity

    • Wage rates not only cost, but lower productivity may also increase total cost

  • Exchange rates and currency risks

  • Costs

    • Tangible - easily measured costs like utility labor materials taxes

    • Intangible - public transportation etc

  • Political risk, values, and culture

    • National state local gov attitudes toward private and intellectual property

  • Proximity to suppliers

    • Perishable goods, high transportation costs, bulky products

  • Proximity to consumers/market

    • Very important to services

    • JIT systems or high transportation costs may make it important to manufacturers

  • Proximity to competitors

    • Often driven by resources such as natural, info, capital, talent

    • Found in both manufacturing and service industries

• Numerical question about the Factor-Rating Method

  • Calculate all weights time all scores to get weighted score

    • =SUMPRODUCT in Excel (select weight column then comma then score column)

• Numerical question about the Center-of-gravity method

  • FInds location of distribution center that minimizes distribution costs

  • Considers location of markets, volume of goods shipped to those markets, shipping cost (or distance)

  • X-coordinate of the center of gravity = Sum of x × Q / sum of Q

  • Y coordinate of the center of gravity = Sum of y × Q / sum of Q

    • Where x = x coordinate of location i

    • y = y coordinate of location i

    • Q = quantity of goods moved to or from location i

• Center of gravity method:

  • X-coordinate of center of gravity = sumproduct (x-coordinate , volume) / sum (volume)

Exponential smoothing

• New forecast = last period’s forecast + a (last period’s actual demand - last period’s forecast)

  • Ft = Ft-1 + a (At-a - Ft-1)

• Where Ft = new forecast

  • Ft-a = previous period’s forecast

  • a = smoothing (or weighting) constant, 

• Least squares method

  • Equations to calculate the regression variables

  • y^ = a + bx

  • B = (sum of) (xy - nyx) / sum of x^2 

  • 
    

Mean squared error (MSE) = sum of (forecast errors)^2 / n

Mean absolute deviation (MAD) = sum absolute value of error / n

Mean absolute percent error (MAPE) = (( sum of absolute value of error / actual value ) / n ) *100

Weighted moving average = sum of ((weight for period n) × (demand in period n)) / sum of weights. Ex: ((3 × 10) + (2 × 12) + (1 × 8)) / 6