ERP 수요 관리 - mailab.snu.ac.kr

advertisement
지난 주
• 1주차
•
•
•
•
산업공학이란?
ERP/SCM 이란?
Operations Management란?
Innovation의 예
• 2주차
• Demand Management
• 데이타
SNU MAI LAB
0
ERP:
Mfg Plan’g and Control(MPC)에 근거하여
Demand Management
SNU MAI LAB
Contents for demand management
Demand Mgt and MPC Environment
Communication with other MPC Modules
Forecasting Models
Conclusion
SNU MAI LAB
2
MPC Concept
중심에는 Business Plan, Sales and Operation Plan, Master
Production Schedule, Material Requirement Plan이 존재한다. 즉
4단계의 Plan을 통해 생산계획이 이루어진다. MPC(Mfg Planning
and Control) 의 기본적 개념이다.
1. Front End는 MPS까지, 2. Engine은 MRP까지, 3. Back End는
두 개의 PO(Production Order와 Purchase Order)가 존재.
1. Frond End의 좌우에는 Demand(Management)와
Resource(Plan)이 존재
2. MRP엔진의 좌우에 PS(Planning and SchedulingRCCP,DCCP,APS)와 MD(Master Data-BOM,IR,R)가 존재.
그림?
SNU MAI LAB
3
Manufacturing Planning and Control
Business Plan
Front End
Resource Planning
Sales & Operation Plan
Demand Management
Master Production Scheduling
PS
MD
(planning & scheduling)
(master data)
RCCP
Engine
(rough cut capacity planning)
DCP
Material Requirement Planning
(detailed capacity planning)
(inventory record)
(advanced planning & scheduling)
Back End
SNU MAI LAB
B/M
IR
APS
Purchasing Order
Routing
Production Order
Outsourcing
Manufacturing Execution System
[by 이재봉,박진우]
4
MPC Concept
Flow Shop, Repetitive Shop, Job Shop, Project Shop
Lean Manufacturing(JIT: Just In Time)의 적용 범주?
MRP의 적용범주?
SNU MAI LAB
5
Demand Mgt and MPC Environment
Customer Order Decoupling Point
Independent Demand vs. Dependent Demand
MTS(Make To Stock), ATO(Assemble To Order), MTO(Make To
Order) and ETO(Engineer To Order)의 MPC환경이 존재
예: 양복점 또는 피자가게의 MTS, ATO, MTO, ETO
SNU MAI LAB
6
Demand Mgt and MPC Environment: MTS
Final Goods Inventory, How much and when to order
Physical Distribution Considerations:
-Plant Warehouse, Distribution Centers, Local Warehouse
-VMI(Vendor Managed Inventory)
Balancing the Level of Inventory vs. Level of Service
Better Forecast, Rapid Transportation, Speedy and More
Flexible Manufacturing
SNU MAI LAB
7
Demand Mgt and MPC Environment: ATO, MTO, ETO
ATO: Personal Computer, Car, Some Industrial Products
Configuration Management, Modules, Options Components
Inventory Advantage over MTS(예: Computer)
4 processor options, 3 hard disk options,
4 CD-DVD options, 2 speakers, 4 monitors
374 final products vs. 17 components
ETO: 설계 능력 및 설계 용량
SNU MAI LAB
8
Communication with Other Modules: Pyramid Forecasting
Aggregation에 따른
Variance의 변화는?
SNU MAI LAB
9
Forecasting Models
Simple Models vs. Complicated Models
Moving Average, Exponential Smoothing, Holt Winters, HW
Seasonal, …
예측 주체: 비전문가, 마케팅 전문가, 예측전문가
SNU MAI LAB
자료 소스: KAIST 전덕빈 교수
10
Forecasting Models
수요예측 정보의 소스:
1. Data(주로 시계열), 2.상식, 3.지식(소비자에 대한, 그리고
이론지식), 4.경험(영업 담당자), 5. 환경(신상품, 기술혁신,
경쟁, 규제완화, 고객 행태 및 구매력 변화)
모델: 시계열 모델 (Time Series, BJ)
vs. 인과관계 모델(Regression, Econometric, Causal Rel.)
SNU MAI LAB
자료 소스: KAIST 전덕빈 교수
11
Forecasting Models
Time Series Model
The general representation of an autoregressive model, well-known as AR(p), is
where the term εt is the source of randomness and is called white noise. It is assumed to have the following
characteristics:
With these assumptions, the process is specified up to second-order moments and, subject to conditions on the
coefficients, may be second-order stationary.
If the noise also has a normal distribution, it is called normal or Gaussian white noise. In this case, the AR process may be
strictly stationary, again subject to conditions on the coefficients.
Regression Model
In the more general multiple regression model, there are p independent variables:
where xij is the ith observation on the jth independent variable, and where the first independent variable takes the value 1
for all i (so is the regression intercept).
The least squares parameter estimates are obtained from p normal equations. The residual can be written as
The normal equations are
In matrix notation, the normal equations are written as
where the ij element of X is xij, the i element of the column vector Y is yi, and the j element of
n×1, and
is
. Thus X is n×p, Y is
is p×1. The solution is
For a derivation, see linear least squares, and for a numerical example, see linear regression (example).
SNU MAI LAB
12
Conclusion
Supply Chain의 가장 중요한 부분
모델과 실제 경험 부분은 수업 중 강의 내용과 위키피디아 자료 및
별도의 비공개 핸드아웃 참조할 것.
SNU MAI LAB
13
비행기 승객 데이터
SNU MAI LAB
14
Download