unit 1 scm

structural element

• large capital investments that are difficult to reverse

• includes tangible resources like buildings, equipment, computer systems

infrastructural element

the policies, people, decision rules and organizational structure choices made by a firm. affect culture and operation of the business. easier to change comparatively.

structural and infrastructural

each part of a business contains ? and ? elements

mission statement

• reason for existence

• core values

• domain

business strategy

• protect core competency

• targeted customers/markets

• areas of sustainable competitive advantage/core competency

• role of supply chain partners

• time frames and performance objectives

operations and supply chain strategies

• translate business strategy into operations and supply chain actions

• provide value to targeted customers/markets

• develop supporting core competencies in operations and supply chain practices

operations and supply chain strategy

• functional strategy that indicates how structural and infrastructural elements with the operations and supply chain areas will be acquired and developed to support the overall business strategy

3 primary objectives of operations and supply chain strategies

• help management choose the long term right mix of structural and infrastructural elements based on a clear understanding of the performance dimensions valued by customers and the trade offs involved

• ensure that the firm’s structural and infrastructural choices are strategically aligned with the firm’s business straegy

• support the development of core competencies in the firm’s operations and supply chains

structural decision categories

• capacity

• facilities

• tech

infrastructural decision categories

• organization

• sourcing/purchasing

• planning and control

• business processes and quality management

• product and service development

4 major performance dimensions

• quality

  • performance

  • conformance

  • reliability

• flexibility

  • mix

  • changeover (switch to a diff)

  • volume (incr/decr output)

• time

  • delivery speed

  • delivery reliability (deliver ON TIME, sometimes preferable to being early)

• cost

  • labor

  • material

  • engineering

  • quality-related

  • overhead allocation

straddling

seeking to compete on all performance dimensions. may be a very risky strategy

trade offs among performance dimensions

• difficult to excel at all 4 (cost, flexibility, time, quality)

• some common conflicts

  • low cost vs high quality

  • low cost vs flexibility

  • delivery reliability (on time ) vs flexibility

  • conformance quality vs product flexibility

• straddling (want to compete on all 4). risky.

order qualifiers

• organization has to have these in place for customer to even consider buying the product or service

• performance dimension on which customers expect a minimum level of performance

order winners

• characteristics that attracts a customer to that particular product or service

• performance dimension highly valued by customers that differentiates a company’s products and services from its competitors

• tends to drive market share within a targeted market segment

prioritizing dimensions of organizational performance

• ID product dimensions highly valued by customer, select competitive dimensions, and design supply chain to excel on these dimensions

• order winners

  • characteristics that attracts a customer to that particular product or service

  • performance dimension highly valued by customers that differentiates a company’s products and services from its competitors

  • tends to drive market share within a targeted market segment

• order qualifiers

  • organization has to have these in place for customer to even consider buying the product or service

  • performance dimension on which customers expect a minimum level of performance

  could have order winners and not enough order qualifiers and then ur not even considered. sometimes order winners can also regress and become order qualifiers

feedback loop

from functional strategy that feeds back to business what are our capabilities/changes we can make. when we change business strategy we need a functional ability to actually deliver that

stages of alignment btw supply chain and operations strategies

• stage 1

  • internally neutral: not linked to business strategy

  • just try to deliver at beginning, not much strategy

• stage 2

  • externally neutral: follow industry best practices

  • then look at competition and basically copy bc u don’t know what else

• stage 3

  • internally supportive: SC strategy aligned with business strategy

  • make concerted effort to align supply chain and business strategy. supply chain supports business strategy. (most successful companies)

• stage 4

  • externally supportive: develop/exploit SC core competencies

  • world class

  • supply chain essentially becomes ur business strategy.

framework for order winners/qualifiers

• divide mkt into segments based on order winners/qualifiers

  • cost, flexibility, service, quality, etc

• ID current market segment or segment(s) to enter

• translate order winners/qualifiers into process requirements

• design processes to meet requirements

  • equipment, facility, labor, etc

• design infrastructure to support processes

  • information and accounting systems, HR, etc.

structural decisions

capacity, facilities, tech

infrastructural decisions

organization, sourcing, planning & control, business processes, quality management, product development

customer value index

a measure that uses performance and importance scores for the various performance dimensions for a product or service to calculate a score that indicates the overall value of an item or service to a customer

In= importance of dimension n (want high)

Pn performance for dimension n (want high)

• sometimes this is just one component like if the difference is immaterial then may want to consider what other options the less “valuable” supplier offers

• could reconfigure our product to favor a diff supplier

forecast

• an estimate of the future level of some variable

• underlying basis of all business decisions

  • mkt: promotions and sales

  • supply chain: purchasing, capacity, production, inventory

  • finance: cash flow projections, profits

  • HR: hiring/firing

  • MIS: user base size, tech development
    challenge: actual future level of variable likely is higher/lower than the prediction (often significant amt)

demand forecasts

• considering overall mkt demand and firm-level demand

supply forecasts

predict material availability based upon suppliers, trends, risk

price forecasts

• forward buying (know price is low rn but predict will increase. determine that price of storing them is less than addtl cost of buying later)

• futures contracts (pay for materials now but don’t take ownership of them until future. pay lower price. don’t pay for storage)

• buying frequency (price is high, predict will decrease. increase frequency but smaller qty to take advantage of price drops)

benefits of better SC forecasts

• lower inventories

• reduced stockouts

• smoother production plans

• reduced costs

• improved customer service

economic forecasts

• inflation rates, borrowing rates

forecasting laws

• almost always wrong (but still useful)

• short term often more accurate than long term

• ? for groups/categories of products/services tend to be more accurate than ? for specific products or services

  • cheaper to forecast for group than individual (think like DOW)

• not substitute for calculated values. only use ? when more reliable method isn’t available

forecasting steps

1. determine how it will be used

2. select values

3. determine planning time horizon

4. select potential model(s)

5. gather historical data

6. calculate using the model(s)

7. evaluate accuracy and choose model

8. make future predictions based on the model

long range forecast

• asset acquisition

• yearly planning bucket

• 3-10 years planning horizon

• new product planning, facility cosntruction, tech

medium range forecast

• asset utilization

• monthly/quarterly planning bucket

• 3 mo- 2 yrs planning horizon

• seasonal production, inventory, employment, budgeting

short range forecast

• asset execution

• weekly/monthly planning bucket

• 1-26 week planning horizon

• job scheduling, worker assignments, inventory stocking

qualitative forecasting methods

• subjective (opinion) based

• used when there is limited quantitative data available

• used when the relationship btw the past and the future is uncertain

  • new or trendy products

  • new tech

• involves intuition, experience

  • forecasting sales of a new product

• ex., market surveys, panel consensus, delphi method, life cycle analogy method, build up

quantitative methods

• objective (calculation) based

• used when quantitative historical data is available

• used when the relationship btw the past and the future is predictable

  • existing or stable products

  • current technology

• involves mathematical techniques

  • forecasting sales of products within a stable market

market surveys

• structured questionnaires or mkt research panels

• qualitative forecast method

panel consensus forecasting

• experts meet together to develop forecasts

• qualitative forecasting method

delphi method

• experts develop forecasts separately and then revise

  • since separate, less likely to be initially influenced by groupthink

• qualitative forecast method

life cycle analogy method

• modeling growth and decline based upon similar products

• qualitative forecasting

build up forecasts

• market segment experts develop forecasts that are added together

• qualitative forecasting

• basically sales ppl give to regional directors who give to forecast to corporate and then nationwide basically.

id underly demand pattern

key to quantitative forecasting

demand patterns

• key to quantitative forecasting

• level or constant

  • avg value is relatively constant over time

• trend

  • long term movement up or down in a time series

• seasonality

  • a repeated pattern of spikes or drops associated with certain times of the year

• cyclical

  • long term cycles of demand often over several years

    • ex., product life cycles, presidential election impact on markets

• randomness

  • unpredictable movement from one time period to the next

  • often serves to hide underlying demand pattern

  • random variation is what makes forecasts esp difficult

    • measure by “variance” or “standard deviation”

level demand pattern

• avg value is relatively constant over time

• shown w/ random variation

seasonality

• a repeated pattern of spikes or drops associated with certain times of the year

• shown w/ random variation

quantitative forecasting approaches

• time series models

  • demand follows a trend and/or pattern over time

    • last period or naive forecast

    • moving average

    • weighted moving average

    • exponential smoothing

    • adjusted exponential smoothing*

    • linear regression*

• causal models

  • demand predicted by observing environmental factors like economic indicators

    • linear regression*

    • multiple regression*

time series

• quantitative forecasting model

• a set of periodic observations arranged in chronological order

  • period is the regular frequency with which measurements are plotted

• assumption

  • past is good predictor of the future

• risk underlying demand pattern may change over time

• trade offs

  • forecast responsiveness

  • forecast stability

last period/naive approach

• current demand becomes the next period’s forecast

• simplest time series model

• very responsive to demand changes in short term, but unstable for long range planning

• major weakness: does not consider historical trends and patterns over time

moving average model

• forecasts average the n most recent demand values

• smooths data randomness to illuminate data pattern

• can be adjusted for the type of data being measured

  • more responsive: smaller n values (for data with frequent data pattern changes)

  • more stability: larger n values (for data with infrequent data pattern changes)

  • which day are u standing in, what are u forecasting 2. . ex., standing in day 11.

    • ex., period 1, demand 3. n=1 (standing in period 1, 3/1, forecast period 2 is 3)

    • could not have a forecast if not enough to meet n (like if n=5, u need 5 data points, take the 5 most recent points)

• does not consider trend or seasonality

• NOTE: n=1 is the same as the last period forecast

weighted moving avg method

• a form of the moving avg that applies varying weights to past observations

• weights are decimal values btw 0 and 1

• weights are listed, the sum of the weights must equal 1 (check this!!)

  • first weight listed is applied to the most recent historical demand

  • last weight listed is applied to the n-th most recent historical demand

• highest weights usually placed on more recent past demand

• cyclical patterns can be modeled by adjusting weight values

• responsiveness determined by weight values

exponential smoothing model

• a form of moving avg model in which the forecast for the next period is calculated as the weighted avg of the current period’s actual value and forecast

• allows one to adjust the balance btw forecasted values and actual values

• the closer alpha is to 1, the greater the weight put on the most recent actual demand value

• bc this involves utilizing prev forecast, an initial forecast (using some other method) will always be given

  • this involves using prev forecast which is based on prev forecast and so on

  • a forecast based on this method is somewhat influenced by all previous forecasts

exponential smoothing model

• features

  • weighted model with greater weight on most recent data

  • requires very little stored data (only the past period forecast and demand) and easy to automate

• the general rules for determining alpha

  • stability

    • the greater the randomness in the time series data, the lower the alpha value (more weight placed on forecast values)

    • the less randomness in the time series data, the higher the alpha value (more weight that is placed on the most recent demand value)

  • responsiveness

    • greater instability in underlying pattern, the higher the alpha value

    • the less instability in the underlying demand pattern, the lower the alpha value

  • alpha is selected to minimize forecast error

  • ? constant/alpha is typically btw (incl) 0.05 and 0.3

• not well suited to forecast trend or seasonality (tends to lag trending and seasonal data). better for flat or slowly changing data

alternative methods to address trending or seasonality

• adjusted exponential smoothing

  • expanded exponential smoothing calculating considers trending

  • contains an “unadjusted forecast” component (alpha)

  • contains a “trend adjustment” component (beta)

• linear regression

  • attempts to model underlying trends using a straight line

  • form of the line is forecast = constant + slope (time)

• seasonally adjusted linear regression

  • multiplies linear regression forecasts by a seasonal index

adjusted exponential smoothing

• expanded exponential smoothing calculating considers trending

• contains an “unadjusted forecast” component (alpha)

• contains a “trend adjustment” component (beta)

linear regression forecast

• attempts to model underlying trends using a straight line

• form of the line is forecast = constant + slope (time)

• persistent, up or down, generally linear movement in a time series

• trends will change over time

  • increase, decrease, remain cosntant

• demand forecast = a + bx, where

  • a = “intercept” at period 0

  • b= demand increase per period

  • x= desired forecast period

seasonally adjusted linear regression

• multiplies linear regression forecasts by a seasonal index

causal forecasting/correlation models

• quantitative forecasting method that predicts a dependable variable as a mathematical function of something other than time (the independent variable)

  • linear regression

    • establishes a mathematical relation btw one dependent variable and one independent variable

  • multiple regression

    • establishes a mathematical relation btw one dependent variable and multiple independent variables

  • dependent variable

    • the variable that is assumed to be “caused”

  • independent variable

    • the variable(s) that are assumed to do the “causing”

linear regression

• causal forecasting/correlation model

• establishes a mathematical relation btw one dependent variable and one independent variable

multiple regression

• causal forecasting/correlation model

• establishes a mathematical relation btw one dependent variable and multiple independent variables

measures of forecast accuracy

• used to assess how well a model is performing or to compare multiple forecast models to one another

• closer to 0 is better

  • forecast error

    • difference btw forecast and actual demand

  • running sum of forecast error

    • indicates tendency to over/under forecast

    • positive bias (demand exceeds the forecast over time)

      • forecasts with positive bias will eventually cause stockouts (no supply)

    • negative bias (demand less than forecast over time)

      • forecasts with negative bias will cause excessive inventory

running sum of forecast error RSFE

• indicates tendency to over/under forecast

• positive bias (demand exceeds the forecast over time)

  • forecasts with positive bias will eventually cause stockouts (no supply)

• negative bias (demand less than forecast over time)

  • forecasts with negative bias will cause excessive inventory

• sum of all the errors basically

• can manipulate a positive/negative bias (ex., for positive if fruit will go bad or u know clothing will go out of style, produce less)

forecast error

• difference btw forecast and actual demand