Introduction to Quantitative Analysis (Business Analytics) - Slides PDF

INTRODUCTION TO QUANTITATIVE ANALYSIS (BUSINESS ANALYTICS) ITIS 1P97: DATA ANALYSIS AND BUSINESS MODELLING The slides are adapted from textbook slides provided by the publisher, materials from the textbook and various IS materials; and are for classroom purposes only. Decisions ….. Decisions…...

INTRODUCTION TO QUANTITATIVE ANALYSIS (BUSINESS ANALYTICS) ITIS 1P97: DATA ANALYSIS AND BUSINESS MODELLING The slides are adapted from textbook slides provided by the publisher, materials from the textbook and various IS materials; and are for classroom purposes only. Decisions ….. Decisions….. Decisions  Think about some decisions you had to make recently or even today?  Which program I want to pursue?  Choosing housing/a rental place?  Which car to buy?  Or even what should I have for breakfast today?  And how did you decide on your Quantitative Analysis  A scientific approach to decision-making referred to as Management Science Operations Research Decision Science Business Analytics  The use of data, information technology, statistical analysis, quantitative methods, and mathematical or computer-based models to help managers gain improved insight about their business operations and make better, fact-based decisions.  “a process of transforming data into actions through analysis and insights in the context of organizational decision making and problem solving.” (Liberatore and Luo, 2010)  Business Analytics DATA Pre-Transaction Data Transaction Data (purchase, service) Post Transaction Data Clickstream data – show in detail exactly where a user goes and what they do, from search engine searches to websites visited, conversions made, and purchases carried out.  Important Ingredients and Processes DATA Pre-Transaction INSIGHT PROCESSING Data (ANALYTICS) Transaction Data (purchase, service) Information Post Transaction Technology Data Statistics Clickstream data – show in Computer detail exactly where a user Science goes and what they do, from …. search engine searches to websites visited, conversions made, and purchases carried out. Business Cases… just a few   Product recommendation In the business context think of decision  Amazon, Netflix, Facebook, LinkedIn, Google/YouTube situations…  As much as 75% of Netflix’s streaming video can be attributed to movie Marketing, Sales, Product Development recommendation  Customer segmentation  Finance  Identify segments of customers that behave similarly with regard to their Operations interaction with the business  Predictive  Human maintenance resources  Failure patterns of various types of equipment are predictable  Predictive medical diagnosis ANALYTICS CATEGORY QUANTITATIVE ANALYSIS TECHNIQUE Descriptive analytics: Statistical measures such as the use of data to understand past means and standard and current business performance deviations, probability and make informed decisions distributions What happened? What is happening? Predictive analytics Decision analysis and decision predict the future by examining trees historical data, detecting patterns or Regression models relationships in these data, and then Forecasting extrapolating these relationships Simulation forward in time. What will happen? Why will it happen? Prescriptive analytics: Linear programming identify the best alternatives to Integer programming, goal minimize or maximize some programming, and nonlinear Models in Business Analytics  Model - an abstraction or representation of a real system, idea, or object.  Captures the most important features  A physical replica, a written or verbal description, a visual representation, a mathematical formula, or a spreadsheet. Known parameters: Known parameters: Selling price per unit $20.00 Selling price per unit $20.00 Fixed cost $25,000.00 Fixed cost $32,000.00 Variable cost per unit $12.00 Variable cost per unit $10.00 Input Data Input Data Number of units, X $1,000.00 Number of units, X 1000 Results Results Total revenue $20,000.00 Total revenue $20,000.00 Fixed cost $25,000.00 Fixed cost $32,000.00 Total variable cost $12,000.00 Total variable cost $10,000.00 Total cost $37,000.00 Total cost $42,000.00 Profit ($17,000.00) Profit ($22,000.00) Break Even Point BEP Number of units, X 3125 BEP Number of units, X 3200 Useful diagrams  Influence Diagrams  Decision Trees  Flow charts  Network Models/Diagrams Decision Models  Decision model - a logical or mathematical representation of a problem or business situation that can be used to understand, analyze, or facilitate making a decision.  Inputs:  Data, which are assumed to be constant for purposes of the model.  Uncontrollable variables, which are quantities that can change but cannot be directly controlled by the decision maker.  Decision variables, which are controllable and can be selected at the discretion of the decision maker.  Variables vs Components of a mathematical model Constants Parameters (uncontrollable factors or Environmental factors) Input or Dependent Independent or Mathematic variables or decision variables al Outputs (controllable relationships ( Projected factors) Results) Problems in which the objective is to find the best solution with respect to one criterion are referred to as single-criterion decision problems. Problems that involve more than one criterion are referred to as  Variables Mathematical vs Models Constants  Deterministic Model – if all uncontrollable inputs to the model are known and cannot vary  Stochastic (or Probabilistic) Model – if any uncontrollable are uncertain and subject to variation  Stochastic models are often more difficult to analyze.  In our simple production example, if the number of hours of production time per unit could vary from 3 to 6 hours depending on the quality of the raw material, the model would be stochastic.  Cost/benefit considerations must be made in selecting an appropriate mathematical model.  Frequently a less complicated (and perhaps less precise) model is more appropriate than a more complex and accurate one, due to cost and ease of solution considerations. Developing a Quantitative Analysis Model A mathematical model of profit Profit = Revenue – Expenses Revenue and expenses can be expressed in different ways  Prichett’s Precious Time Pieces The company buys, sells, and repairs old clocks The parameters Rebuilt springs sell for $8 per unit of this model are Fixed cost of equipment to build springs is … $1,000 Variable cost for spring material is $3 per unit The decision variable of interest is … Write a mathematical model to express profits for Profit can be this problem expressed as … s = selling price f = fixed costs v = Companies are variable costs often interested Number of spring sets sold = X in the break-even point (BEP). It can be What would be profit if 0 items are sold? expressed as What would be profits if 1000 items are sold? … Mathematical Models Profit = Revenue – Costs Profit = Revenue – (Fixed cost + Variable cost) Profit = (Selling price per unit)(Number of units sold) – [Fixed cost + (Variable costs per unit)(Number of units sold)] Profit =sX – [f + vX] Profit =sX – f – vX where s = selling price per unit v = variable cost per unit Prichett’s Precious Time Pieces The company buys, sells, and repairs old clocks Rebuilt springs sell for $8 per unit Fixed cost of equipment to build springs is $1,000 s=8 Variable f = 1,000 v = 3 cost for spring material is $3 per unit Number of spring sets sold = X Profits = Revenue – Costs = sX – [f + vX] = sX – f - vX Profits = $8X – $1,000 – $3X If sales = 0, profits = – f = –$1,000 If sales = 1,000, profits= [($8)(1,000) – $1,000 – ($3)(1,000)] = $4,000 Prichett’s Precious Time Pieces The company buys, sells, and repairs old clocks Rebuilt springs sell for $8 per unit Fixed cost of equipment to build springs is $1,000 Variable cost for spring material is $3 per unit s = selling price f = fixed costs v = variable costs Number of spring sets sold = X Companies are often interested in the break-even point (BEP), What is the break even point? Write a mathematical model to express breakeven point. Pritchett’s Precious Time Pieces Companies are often interested in the break-even point (BEP), BEP is the number of units sold that will result in $0 profit 0 = sX – f – vX, or 0 = (s – v)X – f BEP for Pritchett’s Precious Time Pieces Solving for X, we have BEP = $1,000/($8 – $3) = 200 units f = (s – v)X Sales of less than 200 units of rebuilt springs will result in a f loss X= Sales of over 200 units of s–v rebuilt springs will result in a profit Fixed cost BEP = (Selling price per unit) – (Variable cost per unit) Running Case example  Brock Club (a student club) engages is providing a variety of programs, opportunities, and services for students. To raise funds for the student club, GoodSBSA sells various products (like T-shirts, jerseys etc.) through their shop, runs cafeterias, provides tutoring services, and many other services. PractPr_ Problem 1  Brock Club (a student club) sells print jerseys as one of the product offerings and considering purchasing printing equipment. Two vendors have presented proposals. The fixed cost is $75, 000 for proposal A and $85,000 for proposal B. The variable cost is $34 for A, and $27 for B. The revenue generated by each unit is $75. A. If the expected volume is 1500 units, which alternative should be chosen? B. If the expected volume is 3500 units, which alternative should be taken? C. What is the BEP in units for proposal A? D. What is the BEP in units for proposal B? E. What is the BEP in dollars for proposal A? F. What is the BEP in dollars for proposal B? Problem Solving With Analytics Problems exist when there is a gap between what is Recognizing a problem happening and what we think should be happening. Eg., For logistics company - costs are too high compared Finding thewith real competitors. problem and distinguishing it from Defining the Problem symptoms that are observed. Costs can be higher for many reasons! inefficiencies in routing trucks; OR poor locations of distribution Stating goals centers; OR fuel costs and objectives; Characterizing the Structuring the Problem (related solutions – improving routing process, oror possible decisions; Identifying any constraints redesigning restrictionsentire distribution network, or negotiating better fuel purchases) Involves some sort of experimentation or solution redesigning process, such entire distribution as evaluating network different => new scenarios, Developing a Solution/ Not a trivial task location ofrisks – complexity manufacturing can increase (large analyzing associated plant and distribution with various decision Analyzing the Problem number centersof(Where) courses of action; competing objectives; alternatives, finding a solution that meets certain time OR limitations, different etc…) to which plants and which products goals, or determining an optimal solution. centers? (which) Models cannot capture every detail of the real OR quantities from different plants and centers (how Interpretation problem much) Managers must understand the limitations of models and their underlying assumptions and often Implementing the Solution incorporate judgment into making a decision. Requires providing adequate resources, motivating Implementation employees, eliminating resistance to change, modifying organizational policies, and developing Note: Iterative and incremental trust.process depending on size and Why use mathematical models?  They force us to think logically  They let us use and understand data – to give us insight and information  Models can save time and money in decision making and problem solving  A model may be the only way to solve large or complex problems in a timely fashion  Single stage learning and Two stage learning Models are realistic, solvable, and understandable mathematical representations of a situation  Generally, experimenting with models (compared to experimenting with the real situation): $ Sales b 1X b0 +  requires less time Y=  is less expensive $ Advertising Impacts and Challenges  Benefits  …reduced costs, better risk management, faster decisions, better productivity and enhanced bottom-line performance such as profitability and customer satisfaction.  Challenges  …lack of understanding of how to use analytics,  competing business priorities,  insufficient analytical skills,  difficulty in getting good data and sharing information, and  not understanding the benefits versus perceived costs of analytics studies. Impacts and Challenges  All models are wrong… but some can be useful British statistician George Box