scm 453x Exam 1 – Iowa State University

MTS Environment (Make-To-Stock)
Focus on maintaining finished goods inventory
—When, how much, and how to replenish stock

Customer service based on in-stock
—Trade-off between inventory and level of service

Improvements are made by
—Improving forecasting
—Decreasing lead time
—Increasing flexibility

ATO Environments (Assemble-to-order)
Configuration management becomes key
—Understanding how components combine to become finished goods
—Huge number of possible configurations, many of which will never be ordered

Focus on inventory of components
—Easier than inventory of finished goods

Must manage customer order process
—Lead times, availability, etc.

ETO/MTO (Make/Engineer-to-order)
Engineering becomes a resource to be managed

Order decoupling point may move to the supplier
—Coordination becomes essentiale

is a manufacturing process in which manufacturing starts only after a customer’s order is received

Demand Management and SOP (sales operations planning)
Demand Management provides forecasts to the SOP process

SOP combines forecasts with resource planning to develop aggregate production plans

Replacing Forecasts With Knowledge
This situation reflects best practices in supplier relationships – customers provide suppliers with their longer-term production plans, not just current purchase orders. This allows suppliers to better plan their operations

Provide suppliers with long-term production plans, not just current purchase orders.
Suppliers can better plan their operations
Must overcome trust issues
Find ways to increase overall supply chain value

Forecasting – Quantitative
Quantitative Methods
Used when situation is ‘stable’ & historical data exist
Existing products
Current technologies
Involves mathematical models
Forecasting – Qualitative
Qualitative Methods
Used when situation is vague & little data exist
New products
New technologies
Involves
Intuition
Experience
Judgment
Types of Forecasting Models
Associative
—Causal relationships
—When demand is driven by changes in other factors

Time Series
—Identify historical patterns and forecast into the future.
—If we know that sales are 20% above average each January, the forecast for next January should be upward 20%.
—This is an inappropriate method for weekly sales fluctuations that are the result of price and advertising changes.

4 components of time series demand
Level
Trend
Seasonality
Error
Simple Moving Average
– Choice of N in the forecast
Large N
Advantages: Smoothes out random variations in demand
Disadvantage: reacts slowly to changes in demand

Small N
Advantages: Reacts quickly to shifts in demand
Disadvantages: Doesn’t smooth out random variation

Exponential Smoothing
Averaging method
Weights most recent data more strongly
Reacts more to recent changes
Widely used, trusted method
Trend Component
Uses beta weight
Trend changes from period to period
To Start Double Exponential Smoothing
Set Level = Demand
Set Trend = 0
No Forecast for period 1
P2 Forecast = Level from P1 + Trend P1
Triple Exponential Smoothing
Need a whole cycle to initialize the model
MAPE
absolute error / actual