Managerial Computing

Prof. Umesh Hodeghatta Rao Xavier Institute of Management Bhubaneswar

Course Logistics

Building models and making decisions

This course is about building models and making decisions
About organising information Being able to ask what-if questions Applying powerful mathematical models

Topics Covered
Spreadsheets Graphs/Lines/Charts Analysing stocks from money.com WHAT-IF-Analysis Marketing Models
Moving Average Exponential Smoothing

Data Warehousing Linear Programming Exploring data using Statistics

Course Agenda
70 - 80% Modeling 20 - 30 % Technology Discussion Emerging Technologies, Guest Lecture, etc

Class Hours
Check AIS for timings Social Media Course Facebook page July 17th Week August 24th Week

Exams and Grading

Mid Term/Quiz 30 % (online) Final Exam 40 % Class Participation 10% Class Project and Presentation 20% Attendance 50% !!

Notes, Text Books

No single book All notes will be on internet/AIS Bring your Laptop to every class
MS Excel is installed (2003/2007)

Business Modeling

Model
Everyone uses models to make decisions. Types of models:
Mental (arranging furniture) Visual (blueprints, road maps) Physical/Scale (aerodynamics, buildings) Mathematical (what youll be studying)

To buy a CAR?

EXAMPLE

Example of a decision Should I buy Mercedes Benz?

Organising Information:
How much money I have? How much money a Merk costs? Do I need to take loan? Do I enjoy driving in highway? Is it safe to drive on highway? Do I like attention of people?

What-if?
What-if questions
What if I can convince my wife to give me money? What if I decide to drive? What If I buy Cesna?

Give one real-life example?

REAL-LIFE EXAMPLE

Real-life
Merril Lynch
5.5 million customers Over 10,000 financial advisors Developed a model to design product features and pricing options Benefits:
$100 million increase in annual revenue $20 billion increase in net assets

 CNBC
Determine program schedules Schedules must meet advertisers demographic and cost requirements Developed optimization model to determine optimal timing and pricing of commercials Benefits:
$100 million increase in annual revenue

Simple Business System

INPUTS
Material, Equipment, Capital, Work Force

PROCESSES

OUTPUT
(Finished Goods and Services)

Feedback Loop

Feedback Loop

Business Operation
Managing Director

Marketing

Production

HR

Finance

Marketing Research other systems Operations Other Systems Other systems

MRP System
Bill of Material (BOM)

Material Production Schedule

Inventory Details

Material Requirement s Planning(MRP)

Order schedule Inventory Updates Due date changes

Bicycle
Frame Assembly
Chassis (1) Seat (1) Handle(1)

Wheel Assembly
Rim(1) Spokes (25) Tyre(1)

Processes
Marketing
Forecasting Market Research

Finance
Payroll Costing Assets and Liabilities Investment

Production
Operations Material Management Inventory Control

Human Resources
Recruitment Performance Training

Project Management
Planning Milestones Quality

DEFINE MODEL?

What is a Model
A model is a representation or abstraction of a real-world object, process, concept or problem which is reduced in scope or complexity relative to the problem itself but yet retains the certain essential aspects which define the particular real-world problem

Why Use Model?

Success of an Organization depends on the managerial decision making
Model facilitates to make better decision
Decision Making

Models + Managerial Judgement = Organization Success When you dont use model
Intuition approach

Using Model
A model may be used to:
Predict Optimize

Decision Model Components

Inputs Model
Outputs

Decision Variables & Parameters

Relationships

Performance Measures or Objective Functions

25

Classification of Models
Deterministic Models
Model components and relevant data are known with certainty
Forecasting, Decision analysis, optimization

Probabilistic (Stochastic) Models

Some components or data are not known with certainty
Monte Carlo simulation, Scheduling and queueing

26

General Modeling Process

Step1 : Diagnose problem Step 2: Organize facts Step 3: Select methodology Step 4: Formulate model Step 5: Solve model Step 6: Interpret results Step 7: Validate Step 8: Optimize Step 9: Implement solution Step 10: Monitor results
27

Basic Modeling Process

Real World Problem
Study model behavior

No

Model valid?

Yes Make decisions

Abstraction
-Variables -Constraints

Model

Model Output

Monitor results

28

Advantages of Using Models

Models allow managers to ask what-if questions
Models force a consistent and systematic approach to the analysis of problems

29

Disadvantages of Models
Expensive and time-consuming to develop and test Often misused and misunderstood because of their mathematical complexity
Assumptions that can oversimplify the realworld system
30

Tools and Techniques

Tools and techniques

We will be using Microsoft Excel
Basic Excel Functions Financial Modeling
NPV(Net Present Value) Investment Appraisal Portfolio Management

Marketing models
Salesforce models Forecasting time series and exponential smoothing Forecasting exponential smoothing

Tools
Models for production operations
Logistics models Production planning and scheduling Queuing models Inventory models with constraints Probability distributions Hypothesis testing checking out a claim! Analysis of variance (ANOVA) Statistical process control

Statistical applications in quality control

Other Tools
SAS Analytics
Provide an integrated environment for predictive and descriptive modeling, data mining, text analytics, forecasting, optimization, simulation, experimental design

SPSS Business Intelligence Tools

Techniques
1. Make a mathematical model 2. Implement it in excel 3. Find out how the answers depend on the input variables 4. Use the inbuilt mathematical functions to do analyses 5. Use the excel graphic packages to make diagrams

How will you learn this?

Option 1: You will not listen to me and stay in your hostel room
Option 2: You will listen to me, go to the labs, and not think about the subject Option 3: You will listen to me and do everything I tell you to

Lab Sessions
Fill out the student questionnaire

Explore Simple Math Functions

Date & Time Average

Exercise 1
You are given a data set which has MIS grades of XIMB-PGDM students. The final grade is calculated based on various components quiz, mid-ter, final term and term project. Your job is to analyze the data to find out:
Percentage of students got A+ and A grades Find out what is the cut off total points

Exercise 2(order/supply)
You run a company that manufactures aircraft components. You have many competitors who are trying to offer lower prices and better service to customers, and you are trying to determine if you can benefit from better supply chain management. The fields in the spreadsheet file include vendor name, vendor identification number, purchasers order number, item identification number and item description (for each item ordered from the vendor), cost per item, number of units of the item ordered, total cost of each order, vendors accounts payable terms, promised shipping date, promised transit time, and actual arrival date for each order.

Prepare a recommendation of how you can use the data in this spreadsheet database to improve your decisions about selecting suppliers. Some criteria to consider for identifying preferred suppliers include the suppliers track record for on-time deliveries, suppliers offering the best accounts payable terms, and suppliers offering lower pricing when the same item can be provided by multiple suppliers. Prepare a report to support your recommendations.

Flight Delay
How many flights ONTIME How many flights delayed Maximum delay is from which Airport What is the Maximum delay time?

41

Exercise 4(T-Shirt Demand)

Mr. Saket Panda is a T-shirt vendor. The fixed cost of any order is INR7500, the variable cost is INR 600 per shirt. Pandas selling price is INR 1000 per shirt, until a week after the IPL tournament when it will drop to INR 600 a piece. The expected demand at full price is 1500 shirts. He wants to build a spreadsheet model that will let him experiment with the uncertain demand and his order quantity.

 In this model the profit is calculated as: Profit = Revenue Cost

The formula would then read =-Fixed_order_cost-Variable_cost*Order + IF(Demand > Order, Selling_price*Order, selling price*Demand+Salvage_value* (Order-Demand)

End Of Session