Business Analytics for professional People

What is Business Analysis from AI perspective and showcasing

1. Business Analytics Creating frontiers for Business

2. Course Duration: 3 Days 8 hours per day Participants: 18-20 Course Outline: Day 1 : Introduction to Business Analysis Day 2 : Learning the Tools for Business Analysis Day 3: Understanding the Process of Business Analysis

3. Day 1

4. Nol card transactions Ticket machines 01 Bus-/Metro and Taxi vehicle tracking 02 RTA app, S'hail , real-time feedback. 04 Surveys Complaints Reviews (Call Centers, Digital) 05 Passenger Counters Gate Sensors Escalator Activity 03 Expo Events Dubai Run Public Holidays 06 Ticketing Systems GPS & Telematics Mobile Apps Customer Feedback Sensors External Calendars 01 02 03 04 05 06 Together, these sources create a 360° view of operational performance and customer behavior in RTA services. Transportation Data Sources

5. 🔐 Data Governance Accurate Secure Timely Documented C onsistent 3. Timely 2. C onsistent Formats and definitions are standardized across systems and departments. Enables seamless integration and reliable comparisons. Data is updated regularly. Ensures relevance for real-time and strategic decisions. 1. Accurate Data is correct, validated, and free from errors or inconsistencies. 5. Documented 4. Secure Protected from unauthorized access. Ensures compliance with privacy laws and internal policies. Ownership, data sources, and definitions are clearly recorded. Supports traceability, auditing, and informed usage Data Governance 🧠 Example: ❌ A missing GPS timestamp → inaccurate ETA → poor rider experience ✅ Clean trip records → better traffic forecasting and resource planning

6. Data Preparation & Cleaning Techniques Combine data from multiple sources (e.g., Nol + GPS + feedback) Consolidate Remove duplicates Handle missing or inaccurate values Correct entry errors (e.g., route codes, timestamps) Clean • Keep relevant periods (e.g., past 90 days) • Exclude test data or cancelled trips Filter • Check logic rules (e.g., trip start time < end time) • Detect anomalies (e.g., 500+ passengers in 1-minute window) Validate • Standardize formats (e.g., dates, station names) • Create calculated columns (e.g., delay duration) Transform “Data is like garbage, You'd better know what you are going to do with it before you collect it” Mark Twain

7. KPI Dashboards & Visualization Dashboards transform transportation data into quick, actionable insights . The Right KPIs, when visualized correctly, will improve service delivery, efficiency, and customer satisfaction. “Good KPIs tell the story in seconds. Great dashboards help teams act on it instantly.”

8. Show trends over time Ridership by hour/day Compare across categories On-time rate across bus routes Show part-to-whole (limited categories) Passenger type distribution (Adult/Student) Show volume + trend together Passenger flow through stations Display a single metric clearly Avg wait time today Show intensity across locations Complaint volume per station Show correlation between 2 variables Delay vs. distance traveled Show parts of a total and compare across categories. Ridership split by type (regular, senior, student) across metro lines Choosing the Right Chart Type Before building dashboards, it's important to match each KPI with the best chart type The wrong visual can confuse — the right one reveals patterns instantly. Show parts of a total and compare

9. What is Business Analysis? How to improve business Using Analytics for improvement Gaining insights from business Continuous improvement

10. Individual Activity

11. SCRS

12. Individual Activity

13. PESTLE P – Political E – Economic S – Social T – Technology L – Legal E – Environment

14. Group Activity

15. Porter’s 5 Force Bargaining Power of Customers Buyer Concentration Buyer Volume Buyer Switching Costs Buyer Information Ability to integrate backward Substitute products Price/ Total Purchases Product Differences Brand Identity Impact of Quality/ Performance Buyer Profits Threat Of New Entry Economies of Scale Proprietary Product Differences Brand Identity Switching Costs Capital Requirements Access to distribution Absolute Cost Advantages Government Policy Expected Retaliation Bargaining Power of Suppliers Differentiation of inputs Switching Costs Presence of substitute inputs Supplier Concentration Importance of Volume to supplier Cost Relative to Total Purchase Impact of inputs on cost or difference Threat of forward integration Threat of Substitutes Differentiation of inputs Relative Price Performance of Substitutes Switching Costs Buyer Propensity to substitute Existing Competitors Industry Growth Fixed Costs/ Value Add Over Capacity Product Differences Brand Identity Switching costs Concentration and Balance Informational Complexity Diversity of Competitors Corporate Stakes Exit Barriers

16. Data & Analytics What is data? How is data related to business? Real Life Analytics Understanding the data

17. Data Information in raw or unorganized form (such as alphabets, numbers, or symbols) that refer to, or represent, conditions, ideas, or objects. Data is limitless and present everywhere in the universe. Reference: http://www.businessdictionary.com/definition/data.html

18. Edward Deming quote 'In God we trust, everyone else bring data"

19. Data and business business needs valuable data and insights Understanding your target audience and customers preferences Data Analytics is a combination of all the processes and tools related to utilizing and managing large/Small data sets Integrating physical and digital shopping spheres Assessing Trends and measures to reduce deviations

20. Data and business Improved Service Level Performance Better Order Fulfillment Improved Supplier Management Maximise Customer Value Driving Down Costs Improved Advertising Better Product Management

21. Data data ‘ deɪtə ’ noun the quantities, characters, or symbols on which operations are performed by a computer, which may be stored and transmitted in the form of electrical signals and recorded on magnetic, optical, or mechanical recording media.

22. Understanding data Unbiased Data Achieve a large sample Ask the right questions (Quantitative & Qualitative) Interpret the data correctly Consider Margin of Error Create a Data Checklist Use Statistical Analysis

23. Activity (25 Minutes) Objective: To collect data to benefit the learning of Business Analytics Target Group: Attendees of the workshop Team Size: 4-5 Members

24. Video The Blind Men and the Elephant

25. Data Modeling Types of Data Models Data Modeling Data Modeling Process Constructing Data Models

26. Types of Data models

27. Hierarchical data model

28. Relational Model

29. Network model

30. object oriented model

31. Object-relational model

32. Data Modeling Data Modeling is a method of defining and analyzing data requirements needed to support the business functions of an enterprise. Data modeling is the act of exploring, understanding and designing data-oriented structures. You identify entity types their purpose and then relationships among them. Structuring the data to understand it’s value Pictorial and visual aides to help in the data framing

33. Exercise (1 hour) Create your data base on the models shared Object is to learn the patterns of the persons who are attending this course What is the objective of collecting the data? What are the data points you will collect? What is the interpretation of the data collected?

34. Points to be noted Identify entities and classify them into their types Identify attributes of each entity Define and apply naming conventions Create standards for commonly used attributes Identifying relationships among entities Applying data model patterns Create and assigning key attributes Perform Normalization to maintain data integrity and reduce redundancy De-normalizing to improve performance

35. Day 2

36. Analytics Descriptive analytics – Past Data Predictive analytics – Future data Prescriptive analytics – Past and Future data Diagnostic analytics – Root Cause Analysis

37. Predictive Analytics

38. Correlation The correlation is one of the most common and most useful statistics. A correlation is a single number that describes the degree of relationship between two variables. Let's work through an example to show you how this statistic is computed. Seek the height of each individual in your class and correlate the data with a histrogram .

39. Regression Regression  analysis is a statistical process for estimating the relationships among variables. It includes many techniques for modeling and analyzing several variables, when the focus is on the relationship between a dependent variable and one or more independent variables (or 'predictors').

40. Scatter Plots Scatter plots are used to establish relationship between two characteristics. The Factor which supposed to determine the value of other factor is called Independent variable. Factor which is assumed to be impacted by the other factor is called Dependent variable. In the equation, y = f (x). X is assiumed to affect the value of y. ie ., y is depending on x Hence, x is the independent variable and y is dependent variable. Always plot independent variable in x axis . In this example, each dot shows one person's weight versus his height.

41. Relationships

42. Regression Analysis – Linear Model A linear relationship simply means that a change of a given size in x produces a proportionate change in y. Described by the line equation y = a + bx

43. Regression - Exercise A Gold ornaments selling company is interested in increasing the sales of it’s outlet and cost is a concern. The company has collected data from it’s past sales on the number of products sold and period it sold it in. The company is looking to increase their sales in the coming year. Chart the data points and data to be captured from the customers to improve the sales based on past data.

44. Multiple R.  This is the correlation coefficient. It tells you how strong the linear relationship is. For example, a value of 1 means a perfect positive relationship and a value of zero means no relationship at all. It is the square root of r squared (see #2). R squared . This is r 2 , the Coefficient of Determination. It tells you how many points fall on the regression line. for example, 80% means that 80% of the variation of y-values around the mean are explained by the x-values. In other words, 80% of the values fit the model. Adjusted R square.  The adjusted R-square adjusts for the number of terms in a model. You’ll want to use this instead of #2 if you have more than one x variable. Standard Error of the regression:  An estimate of the standard deviation of the error μ. This is  not  the same as the standard error in descriptive statistics! The standard error of the regression is the precision that the regression coefficient is measured; if the coefficient is large compared to the standard error, then the coefficient is probably different from 0. Observations . Number of observations in the sample.

45. EXPLAINED PART TWO: ANOVA SS = Sum of Squares. Regression MS = Regression SS / Regression degrees of freedom. Residual MS = mean squared error (Residual SS / Residual degrees of freedom). F: Overall F test for the null hypothesis. Significance F: The significance associated P-Value.

46. INTERPRET REGRESSION COEFFICIENTS Coefficient: Gives you the least squares estimate. Standard Error: the least squares estimate of the standard error. T Statistic: The T Statistic for the null hypothesis vs. the alternate hypothesis. P Value: Gives you the p-value for the hypothesis test. Lower 95%: The lower boundary for the confidence interval. Upper 95%: The upper boundary for the confidence interval.

47. Time Series Forecasting An ordered sequence of values of a variable at equally spaced time intervals Can be used for: Sales Forecasting Budgetary Analysis Stock Market Analysis Yield Projections Process and Quality Control Inventory Studies Workload Projections

48. https://www.youtube.com/watch?v=gHdYEZA50KE

49. Standard Deviation

50. Hypothesis Testing Hypothesis testing is a statistical procedure in which a choice is made between a null hypothesis and an alternative hypothesis based on information in a sample. The null hypothesis, denoted H0, is the statement about the population parameter that is assumed to be true unless there is convincing evidence to the contrary. The alternative hypothesis, denoted Ha, is a statement about the population parameter that is contradictory to the null hypothesis, and is accepted as true only if there is convincing evidence in favor of it.

51. Hypothesis Testing Null hypothesis stands true until and otherwise proved wrong. Accused is considered as Innocent until and otherwise proven guilty.

52. Hypothesis Testing We want to take a practical problem and change it to a statistical problem We use relatively small samples to estimate population parameters There is always a chance that we can select a “weird” sample Sample may not represent a “typical” set of observations Inferential statistics allows us to estimate the probability of getting a “weird” sample

53. Why do Hypothesis Testing? To determine whether there is a difference between processes To validate process improvements (to prove that the process improvements done are yielding results) To identify the factors which impact the Mean or Standard deviation

54. Errors Type-1 (a) Error: Rejecting the Null hypothesis when it is true a-value indicate the probability of Type-1 error happening Type-2 (b) Error: Failing to reject Null hypothesis when it is false b-value indicates the probability of Type-2 error happening

55. Hypothesis & Risk When accepting or rejecting a hypothesis, we do so with a known degree of risk and confidence To do so, we specify in advance of the investigation the magnitude of decision risk and test sensitivity which is acceptable

56. Hypothesis and Decision Risk At  level of significance, the degree of confidence in our decision Is (1- ) which is called confidence coefficient.

57. Prescriptive Analytics

58. FMEA Failures are any errors or defects, especially ones that affect the customer, and can be potential or actual. Failure Mode is the ways or modes by which something might fail. Effect Analysis refers to studying the consequences of those failures

59. Steps to construct FMEA For each of failure modes identified, give scoring (from 1 to 10) based on below table Category Criteria Question Highest Score (10) Lowest Score (1) Severity How severe are the consequences of this failure mode? Fatal / very high No risk Occurrence How frequently it happens? Always Never / Very rare Detection How quickly the failure could be identified by our system? We cant identify, we will come to know this upon consumption by customer The failure will be detected as it happens and will not move further

60. Failure Mode & Effects Analysis Process / product FMEA Date (Original) FMEA Team (Revised) Project Name Page: Of Process Actions Results Process steps / requirements Potential failure mode Potential effects of failure Severity Potential cause(s) of failure Occurrence Current controls Detection Risk priority number Recommended action Responsibility and target completion date Action taken Severity Occurrence Detection Risk priority number Text Text Text Number Text Number Text Number calculation Text Text Text Number Number Number Calculation

61. Cause and Effect Diagram This is also known as Ishikawa Diagram or Fish-bone Diagram. Dr Ishikawa suggested that every problem we face in our shop floor could be caused by any of these six factors, viz., Machine, Method, Material, Measurement, Man and Mother Nature. It is not recommended to add a seventh factor. Machine Method Material Measurement Man Mother Nature 6 Category of causes Effect Poor Process Capability

62. Cause and Effect Diagram In true sense, Fishbone diagram is a stand alone tool for root cause analysis, where brainstorming is the start point, followed as category brainstorming. Root cause analysis is done by drawing branches to the main 6 bone of fish. But in DMAIC, we spilt the functions as Brainstorming or FMEA Fish-bone for categorical analysis and Root Cause analysis using Why-Why Analysis Hence, after brainstorming, fit every primary cause in appropriate category on fishbone diagram and ensure all the 6 bones of the fish is having adequate flesh (number of causes)

63. Root Cause Analysis using Why-Why Why-Why Analysis is a simple and effective method of finding out root causes of a problem. It is a method of questioning all what we know about the problem with why (all potential causes we identified in brainstorming or FMEA). This relentless questioning opens up our thinking process and lead to root cause of the problem. Each of the possible causes listed in the brainstorming or FMEA is questioned and we need to have at least 2 root causes per possible cause.

64. Prioritisation of Root Causes The scoring criteria are as below. Category Criteria Question Highest Score (10) Lowest Score (1) Severity How strongly the root cause and the problem (project y) are connected? Or how much % of problem is associated with this cause? They are strongly related. This root cause give raise to 100% of the problem. Seemingly no connection Occurrence How frequently it happens? Always Never / Very rare Controllability How quickly and completely the team can take actions on this root cause. (team’s authority, expertise and technology) Team can act right away. They are adequately authorised and skilful. Team has to cross multiple approvals and also needs additional expertise

65. Day 3

66. Test Assumptions

67. Analyzing Quantitative Data Frequency Analysis Difference Analysis Qualitative Data Descriptive Analysis Interpretive Analysis

68. Brain Storming SCOT Cause and Effect Comparative Analysis

69. Data Visualization Charts Flow Charts Graphs Pictures Map

70. Software SAS Python Programming R programming Microsoft Excel Google Spreadsheets/ Fusion Tables Tableau (Online tool)

71. MaxStat  by MaxStat Software Complete statistics package with intuitive user interface and easily understandable results. Designed for researchers and students SPSS  by IBM Predictive Analytics can uncover unexpected patterns and associations and develop models to guide front-line interaction Minitab 17  by Minitab Analyze your data and improve your products and services with the leading statistical software used for quality improvement worldwide. DataMelt (" Dmelt ")  by jWork.ORG Data analysis, math and data visuzalization program which combines the power of Python and Java (free) Analytica  by Lumina Decision Systems Analytica is a powerful, stand-alone application for visual quantitative modeling with a full array of statistical analysis functions.

72. Statwing  by Statwing Statwing chooses statistical tests automatically, then reports results in plain English. Statwing is delightful and efficient analysis. Stata  by StataCorp Stata statistical software is a complete, integrated statistical software package. STEM  by Princeton National Surveys Excel macro package for running statistical tests on summary data. Output arranged to easily produce graphs in PowerPoint. XLSTAT  by Addinsoft Variety of tools to enhance the analytical capabilities of Excel, making it the ideal for data analysis and statistics requirements. AcaStat  by AcaStat Software Statistical software and instructional aids to help you quickly organize and analyze data. AlterWind Log Analyzer  by AlterWind A log analyzer tool for determining the basic characteristics of the hits on your site. Analyse-it  by Analyse-it Software Statistical analysis software for researchers in environmental & life sciences, engineering, manufacturing and education. Analysis Studio  by Appricon Provides an end to end model generation process designed for fast development, analysis and deployment.

73. ChemStat  by Starpoint Software Full featured RCRA compliant statistical analysis of ground water data. CoPlot  by CoHort Software A program for making publication-quality maps, technical drawings, and 2D and 3D scientific graphs; includes a statistical add-on. Decision Analyst STATS  by Decision Analyst Windows-based statistical software for marketing research.

74. Decision Science  by Stone Analytics Embeddable analytic engines designed for integration into a wide variety of enterprise applications. Develve  by Develve Statistical Software Statistical software for fast and easy analysis. Basic statistics, Design of Experiments, Gauge R&R and Sample size calculations. EasyFit  by MathWave Technologies Data analysis and simulation software with data management, reporting functionality for probability distribution selection

75. ESBStats  by ESB Consultancy Statistical Analysis and Inference Software Package for Windows. Forecast Pro  by Business Forecast Systems A standalone analytic tool for business forecasting that combines proven statistical methods with an intuitive interface. JMP Statistical Software  by JMP Statistical Software JMP, data analysis software for scientists and engineers, links dynamic data visualization with powerful statistics, on the desktop. KnowledgeSTUDIO  by Angoss Business intelligence and predictive analytics software suite with decision tress and data visualization.

76. MATLAB  by The MathWorks A programming environment for algorithm development, data analysis, visualization, and numerical computation. MedCalc  by MedCalc Software A complete Windows-based statistical program for biomedical researchers. Number Analytics  by Number Analytics Provides statistical analytics software for business users, pricing & promotion optimization, conjoint analysis, new product design. PolyAnalyst  by Megaputer Intelligenc Offers a comprehensive selection of algorithms for automated analysis of text and structured data. Predictive Suite  by Predictive Dynamix Computational intelligence software for data mining analysis and predictive modeling . Scilab  by Scilab Enterprises Open source software for numerical computation providing a computing environment for engineering and scientific applications.

77. SigmaPlot  by Systat Software Systat Software presents award winning scientific data analysis software. The R Project  by R- Project.Org Software environment for statistical computing and graphics (free). Orange  by Biolab.Si Open source data visualization and data analysis for novice and expert (free) Statistix  by Analytical Software Easy-to-use, comprehensive statistics and data manipulation Weka  by Waikato Machine Learning Group Machine learning algorithms for data mining tasks software (free). UNISTAT  by UNISTAT A statistical software package featuring a statistics add-in for Excel data analysis, charting and presentation-quality reporting