Sie sind auf Seite 1von 7

Course Overview

Decision Modeling for Management


Introduction to Decision Modeling

Communications
Email forwarding to Pace account Blackboard

Purpose Prerequisites
Excel and Word basics

Homework Course materials


Textbook
Software

Professor John Carter

Exams Grades Topical scope WSJ subscription: info.wsj.com/student


2

Why Study?
You will develop strong general problem solving skills Your quantitative skills (working productively with numbers) will be sharpened Your Excel knowledge will increase, thus enabling you to apply your problem solving and quantitative skills You will develop your intuition the ability to quickly see through to the essence of a problem
3

What is Decision Modeling?


Decision modeling refers to the use of mathematical or scientific methods to determine an allocation of scarce resources that improves or optimizes the performance of a system Model: A simplified representation of reality
Physical or visual Abstract Needs to be valid to be useful

Decision models are mathematical models


4

Examples of Models
P = (p-v)X - F
Where P = profit p = price per unit v = variable cost per unit X = output volume F = fixed cost
$
1000

EOQ: A Graphical Model

800

Total Cost

600

TC =

400

Carrying Cost Ordering Cost EOQ

This is called the EOQ model and describes the relationship between Total inventory cost (TC) and Demand (D), order Quantity (Q), Ordering cost (OC) and Holding cost (HC)
5

200

0 0 10 20 30 40 50
6

Order Quantity

Typical Management Problems


The following are some examples of management problems that have powerful modeling solutions
Finance - portfolio design, cash flow matching Production production resource allocation, refinery blending Distribution - transportation planning, optimal service routing Marketing - media mix, sales force allocation Risk analysis assessing risk in financial planning System design customer service location and staffing 7

Role of Models
The modeling approach to decision making is best viewed as an adjunct to traditional decision making methods Traditional methods make use of experience, precedent and qualitative analysis of possible action plans In many cases, much better decisions can be made by augmenting this approach with decision modeling
8

The Decision Modeling Process


Step 1: Define the problem
Recognize that you have a problem Identify your objectives in solving the problem Identify the boundaries of the relevant system
Internal vs external Relevant people and organizational units

Example
Blue Ridge Hot Tubs makes two models of tub: AquaSpa and HydroLux Define the problem
The company needs to decide how many to produce for the coming month The objective is to make as much profit as possible The market for both products is strong so any mix will sell The problem context is production planning so doesnt involve marketing, finance, etc
9 10

Process (Continued)
Step 2: Collect data (uses statistics)
Determine what information will be needed and what data would provide that information Assemble or collect relevant data

Process (continued)
Step 3: Build the model
Describe the relationship among all relevant variables controllable (decision), uncontrollable and outcome (affecting objectives) Specify with equations (or inequalities) how the three kinds of variables are related Clearly specify all assumptions and constraints

Application to example
We need data on the following:
Demand (unlimited here) Available pumps, labor hours and feet of tubing Resource requirements for each product Profit contribution for each product
11

Application to example Well discuss in detail the process of building the model shortly, but here is the result:

12

Blue Ridge Data


There are 200 pumps, 1566 hours of labor and 2880 feet of tubing available
Pumps Labor Tubing Unit Profit Aqua-Spa 1 9 hours 12 feet $350 Hydro-Lux 1 6 hours 16 feet $300

Blue Ridge Model


Let X1 = the number of AquaSpa tubs made Let X2 = the number of HydroLux tubs made Maximize 350X1+300X2 Subject to: X1+X2 <= 200 (pumps) 9X1+6X2 <=1566 (labor) 12X1+16X2 <= 2880 (tubing) X1 >= 0 X2 >= 0
14

The Spreadsheet Model

Process (Continued)
Step 4: Verify the model
Show that the model is an accurate representation of the system
Check the logic Check that model response is consistent with known conditions by trying a set of values

Applying to the example


Blue Ridge Hot Tubs Aqua-Spas 1 $350 Hydro-Luxes 1 Total Profit $300 $650 Used 2 15 28 Available 200 1566 2880
16

Number to Make Unit Profits Constraints - Pumps Req'd - Labor Req'd - Tubing Req'd
15

1 9 12

1 6 16

Process (Continued)
Step 5: Solve the model
If the model is analytical, use mathematical analysis (e.g. calculus) or computer programs (e.g. Solver in Excel) to find the optimal solution If the model is purely descriptive, use simulation to explore the behavior of the system under various scenarios

The Solution

17

18

Process (Continued)
Step 6: Communicate the results of the analysis
Provide a clear report on the problem, the assumptions made and the recommendations

Decision Modeling Process


Formulation Real world Decision Model

System

Step 7: Implement the solution


Temper the models solution with other considerations Implement the course of action Monitor progress If necessary, return to the model with updated parameters or relationships and re-solve
19

Implementation Deduction

Interpretation

Real World Conclusions

Model Conclusions

20

Decision Models & Spreadsheets


At a conceptual level, spreadsheet models are very similar to mathematical models:
CARLOAN.XLS

Loan Payment Example

Y = f(X1, X2, , Xn)


Spreadsheet models have cells representing inputs ( Xi) to which mathematical functions ( f (.)) are applied to compute some bottom line output measure (Y)
22

Benefits of Modeling
The modeling process gives us insight and understanding that improves decision making Economy - it is often less costly to evaluate decision options using models Timeliness - models often provide essential information more quickly than their real-world counterparts Feasibility - models can be used to try things that would be impossible in practice

Good Decisions vs. Good Outcomes


Good decisions do not always lead to good outcomes...
If sunny weather is predicted you may (rightly) decide to leave your umbrella at home. If it rains unexpectedly you may get wet (a bad outcome), but that doesnt mean you made a bad decision.

A structured, modeling approach to decision making helps us make good decisions, but cant guarantee good outcomes

Overview of Decision Models


Degree of Uncertainty Single period models incorporating uncertainty Multiperiod Planning models with uncertainty HIGH UNCERTAINTY

Summary
In this introduction, we have discussed the role of decision modeling in management The decision modeling process is a seven step procedure Decision models can be mathematical or in spreadsheets

Static planning models with deterministic assumptions CERTAINTY SINGLE PERIOD MULTIPLE PERIODS

Multiperiod planning models assuming known data

Length of Planning Horizon


25 26

Das könnte Ihnen auch gefallen