Supply Chain exam 2

forecasting

vital function, affects every significant management decision

ex: new product planning, personnel compensation, select suppliers

tactical forecasts

day to day decisions

decoupling point

point at which some form of inventory is stored that allows supply chain partners/entities to operate independently: strategic

influenced by error in forecast and type of product (easily inventoried or perishable)

four types of forecasting

1. time series analysis: relies on future being similar to the past

2. qualitative

3. causal

4. simulation

components of demand

average demand for a period of time, trend, seasonal element, cyclical elements, random variation, autocorrelation

short, medium and long term

short term- forecasting less than three months

medium term-forecasting 3 months-2years

long term- forecasting greater than two years

Forecasting model selection

depends on:

time horizon to be forecast

data availability

accuracy required

size of forecasting budget

availability of qualified personnel

simple moving average (forecast method)

average of a fixed number of past periods

6-12 months, weekly data often used: stationary, short horizon

formula: F_t = (A_t-1 +A_t-2 +…A_t-n)/ n

variable meanings in simple moving average equation

F_t : forecast for period t

n- number of periods to be averaged

A_t-1: actual occurrence in past period

A_t-2: actual occurrence two periods ago

Weighted moving average forecast method

allows unequal weighting of prior time periods, 5-10 observations needed to start

selecting weights: experience and trial and error are best ways, recent past is often best indicator, weights should reflect if data is seasonal or cyclical

Exponential Smoothing

5-10 observations needed to start

weighted average method that includes all past data in the forecasting calculation

more recent results weighted more heavily, most used of all techniques

well accepted because: 1. surprisingly accurate 2. easy 3. user can understand 4.little computation required 5. small computer storage requirements 6. tests for accuracy are easy to compute

Exponential Smoothing Model

F_t = F_t-1 + a(A_t-1 -F_t-1 )

a: adjustment (smoothing constant): correcting trend somewhat, trend in data causes exponential forecast to always lag the actual data

Exponential Smoothing with Trend

FIT_t = F_t +T_t

F_t : exponentially smoothed forecast that does not include trend for period t

T_t : exponentially smoothed trend for period t

FIT_t : forecast including trend for period t

alpha and delta

relatively small values, usually in the range 0.1-0.3

-smoothing constant for random variation and smoothing constant for trend

alpha depends on how much random variation is present, delta depends on how steady the trend is

Linear regression analysis

regression is used to identify the functional relationship between two correlated variables, usually from observed data

assumes the relationship can be explained with a straight line

Y= a +bt

Least Squares Method

determines parameters a and b such that the sum of the squared errors is minimized- “least squares”

Time Series Decomposition

-chronologically ordered data referred to as a time series

time series may contain one or many elements: trend, seasonal, cyclical, autocorrelation, and random

-identifying these elements and separating the time series data into these components is known as decomposition

Seasonal Variation

-may be either additive or multiplicative

seasonal factor (or index) is the ratio of the amount sold during each season divided by the average for all seasons

Decomposition steps

1: determine seasonal factor

2: deseasonalize original data

3: develop a least squares regression line for the deseasonalized data

4: project the regression line through the period to be forecast

5: create final forecast by adjusting the regression line by the seasonal factor

forecast errors

difference between the forecast value and what actually occurred

sources of error: bias (when a consistent mistake is made) random (errors not explained by model being used)

Measures of error: mean absolute deviation: (MAD) mean absolute percent error (MAPE) and tracking signal

forecast error measurements

MAD: ideally zero, means no forecasting error

MAPE: scales the forecast error to the magnitude of demand

MAPE: = MAD/Average demand

Tracking signal: indicates whether forecast errors are accumulating over time (either positive or negative errors)

causal relationship forecasting

uses independent variables other than time to predict future demand-independent variable is a leading indicator

-many apparently causal relationships are merely correlated events

qualitative forecasting

generally used to take advantage of expert knowledge

useful when judgement is required, when products are new, or if the firm has little experience in a new market

ex: market research, panel consensus, historical analogy, delphi method

Collaborative Planning, Forecasting and Replenishment

web-based process used to coordinate the efforts of a supply chain

integrates all members of a supply chain-manufacturers, distributors, and retailers

depends upon the exchange of internal info to provide a more reliable view of demand

CPFR steps

creation of a front end partnership agreement—- joint business planning—-development of demand forecasts—-sharing forecasts—-inventory replenishment

inventory concept

money-on forklifts, shelves, trucks, ships, planes

for many the largest asset on the balance sheet at any given time, can be difficult tho to convert back into cash (liquidity)

good idea to get inventory down as much as possible

inventory models

single period model: used when we are making a one-time purchase of an item

fixed order quantity model: used when we want to maintain an item “in-stock”, and when we restock, certain number of units must be ordered

fixed-time period model: item is ordered at certain intervals of time

inventory definition

the stock of any item or resource used in an org

includes raw materials, finished products, component parts, supplies, and work-in-process

manufacturing inventory

refers to items that contribute to or become part of a firm’s product

inventory system

set of policies and controls that monitor levels of inventory: determines what levels should be maintained, when stock should be replenished, and how large orders should be

inventory costs

holding/carrying costs: costs for storage, handling, insurance

setup (production change) costs: costs for arranging specific equipment setups

ordering costs: costs of placing an order

shortage costs: costs of running out, usually called “stockout”

Demand types

Independent demand: the demands of various items are unrelated to each other

dependent demand: need for any one item is a direct result of the need for some other item (car needs wheel and tires)

types of inventory systems

single-period inventory model: one-time purchasing decision, seeks to balance costs of inventory overstock and understock

multi-period inventory models: 1. fixed-order quantity models, even triggered (running low on stock), 2. fixed time-period models- time-triggered (monthly sales call by sales representative)

stocking level equation

x=u + zo

Known probabilities version

uses a discrete probability distribution

found using data, requires marginal analysis using probabilities

fixed-order quantity model synonyms

also called the economic order quantity, EOQ, and Q-model

event triggered

fixed-time period models

-also called the periodic system, periodic review system, fixed-order interval system, and P-model

time triggered

Economic Order Quantity (EOQ)

Q_opt : occurs where total costs are at their minimum

TC: total cost

Q/2H: holding cost

DC: annual cost of items

D/Q S: ordering cost\

safety stock

refers to the amount of inventory carried in addition to expected demand- common approach is to keep a certain number of weeks of supply, better approach is to use probability

-assume demand is normally distributed, assume we know mean and standard deviation

fixed time period model variables

q: quantity to be ordered

T: number of days between reviews

L: lead time in days

d: forecast average daily demand

z: number of standard deviations required for specified service level

o_T+L : standard deviation of demand during review and lead time

I: current inventory level (including items on order)

inventory turns

number of times inventory is cycled through over time- a measure of how efficiently inventory is used

inventory models with price breaks

-price varies with the order size: to find the lowest cost, calculate the order quantity for each price and see if quantity is feasible

1. sort prices from lowest to highest and calculate order quantity for each price until a feasible order quantity is found

2. if first feasible order quantity is lowest price, this is best we can do

3. otherwise calculate total cost for the first feasible quantity and calculate total cost at each price lower than first feasible order quantity

ABC classification

shows that a small amount of items account for a large dollar volume and large number accounts for small dollar volume

A: top 15% of items

B: next 35%

C: last 50%

inventory accuracy and cycle counting

inventory accuracy: refers to how well the inventory records agree with physical count

cycle counting: a physical inventory-taking technique in which inventory is counted on a frequent basis rather than once or twice a year

logistics definition

the art and science of obtaining, producing, and distributing materials and products in the proper place and in the proper quantities

international: managing these on a global scale

3rd party logistics company

an outside company used to manage all or part of another company’s logistics functions

logistic decision: transportation

truck: great flexibility

ship: high capacity, low cost but slow

plane: fast, expensive, least flexible

train: low cost but slow and variable

pipeline: cheapest, highly specialized and limited to liquids, gases and solids in slurry form

hand delivery: last step in many supply chains

(multimodal solutions are normal)

Warehouse activities

cross docking: large shipments are broken down into small shipments for local delivery in an area, minimizes inventory in the warehouse

Hub-and-spoke system: maintaining as little inventory as possible, sole purpose of warehouse (the hub) is sorting goods to consolidation areas, where each area is designed for shipment to specific location

Facility location issues

1. proximity to customers: makes rapid delivery easier

2. business climate: can include presence of similar-sized businesses, businesses in the same industry, and other foreign companies

3. total costs

4. infrastructure: adequate road, rail air and sea transportation along with energy and telecommunications

5. quality of labor: educational and skill levels must match needs

6. suppliers: proximity of important suppliers supports lean production

7. other facilities

8. free trade zones: a closed facility into which foreign goods can be brought without being subject to normal customer requirements

9. political risk

10. government barriers

11. trading blocs: firms locate within a block to take advantage of new markets or lower total cost

12. environmental regulation:

13. host community: and their interest

14. competitive advantage

plant location methods

factor-rating systems

transportation method of linear programming

centroid method

factor-rating system

-most widely used method

1. list of factors is developed

2. range of possible points is assigned to each factor

3. each site is rated against each factor

4. sums of assigned points for each site are computed

5. site with most points is selected

transportation method

special linear programming method

seeks to minimize costs of shipping n units to m destinations, or it seeks to maximize profit of shipping n units to m destinations

centroid method

used for locating single facilities that considers existing facilities, the distances between them, and the volumes of goods to be shipped between them