MGS 8020 Business Intelligence Analyze Nov 6, 2014

MGS 8020Business IntelligenceAnalyzeNov 6, 2014

Control Improve Analyze Define Measure Analyze Objectives • Determine Variation Sources • Identify internal and external factors and the effect of variability on output • Prioritize Key Outputs and Inputs • Find gaps between current and desired performance and select causes to analyze • Revise Project Benefit • Quantify the project benefit to customers and company

Qualitative vs. Quantitative Analysis Qualitative: Goal - Identify and prioritize critical X’s Tools - Pareto Charts, C&E matrix/fishbone diagram, Brainstorming, Detailed As-Is process, Non-value add analysis, Benchmarking, FMEA Quantitative: Goal – Find Root Causes and Validate critical Xs. Tools - Hypothesis Testing Concepts, Multi-Vari Studies, Simple and Multiple Regression, ANOVA

Agenda • Process Analysis • Qualitative Analysis • Quantitative Analysis

Process Analysis • Process • Any part of an organization that takes inputs and transforms them into outputs repeatedly • Process Analysis • The ability to diagram a process, identify its bottlenecks, and determine its capacity • Process Flowchart • The use of a diagram to present the major elements of a process. These basic elements can include: • tasks or operations • flows of materials or customers • decision points • storage areas or queues • Process terminators

Process Flowchart Example

Multistage Process with Buffer Buffer Stage 2 Stage 1 Multistage Process without Buffer Stage 2 Stage 3 Stage 1 Multistage Processes Buffer A storage area between two tasks where the output of a previous task is placed prior to being used downstream Allows stages to operate independently Processes without Buffer Incur Risk of Blocking - tasks that stop because the next task is not ready to accept Starving - tasks that stop because no work is available

Stage 2 Stage 1 Average WIP = Lead Time (throughp ut time in WIP inven tory) Throughput Rate Buffer Little’s Law Throughput Time Throughput time in WIP • Throughput can be best described as the rate at which a system generates its products / services per unit of time. We often measure their throughput using a mathematical equation known as Little's Law, which is related to inventories and process time: time to fully process a product / service. • Using Little's Law, one can calculate throughput with the equation:I = R * T,where I is the number of units contained within the system, Inventory; T is the time it takes for a unit to go through the process, Flow Time; and R is the rate at which the process is delivering throughput, Flow Rate or Throughput.

Process Analysis with Buffers • A take-out restaurant has a two-step process. The first step is to bake 100 meat loaves. The second step is to individually package each meal. • Baking time is 60 minutes, while it only takes 48 minutes to package the 100 meals. The owner of the restaurant has just purchased a second oven to speed up the process, and since the packaging team was utilized only 80% of the time when only one oven was in operation. • The cycle time for each oven is still one hour for 100 loaves, but with two ovens, the cycle time for the two ovens together is 30 minutes. With two ovens, the packaging process is now the bottleneck activity. The owner would like to balance the process steps, but she cannot afford to purchase any additional ovens at this time, and there is no additional space for ovens. • The Director of Operations has suggested that they could bake for two shifts and package for three shifts. The owner is considering this idea, but is worried that the meat loaves would deteriorate over long wait periods. What do you think?

Process Analysis with Buffers Example of Using Little’s Law Bake 100 meat loaves Baking time 60 minutes Bake 100 meat loaves Baking time 60 minutes CT= 30 min 100/48 x 60 = 125 Individual Pack 100 meat loaves 48 min CT= 48 min

Machine Man Effect Environmental Material Method Cause and Effect Diagram

Pareto ChartsRoot Cause Analysis 80% of theproblems may beattributed to 20%of the causes

Benchmarking • External benchmarking • 1. Identify those processes needing improvement • 2. Identify a firm that is the world leader in performing the process (Library & WWW) • 3. Contact the managers of that company and make a personal visit to interview managers and workers • 4. Analyze data • Internal benchmarking • 1. Identify those processes needing improvement • 2. Identify similar process within the organization • 3. Bring together representatives to share and document best practices • 4. Expand concept to include appreciative inquiry

Benchmarking Phase 1 Phase 2 Phase 3 Phase 4 Phase 5 Phase 6 Internal Benchmarking Reverse Engineering Similar Industry Process Benchmarking Different Industry Process Benchmarking Strategic Benchmarking International Benchmarking Standardize internal best practices Study/ Incorporate best features of competitive products Study/ Incorporate best process practices Study/ Incorporate applicable best process practices Study other approaches to strategy development Leverage core competencies and trends Identify global strategies, customers, new partners, products, and processes

Value Stream Mapping

Overview of the Regression Agenda Statistical Significance Minitab

What is the Regression Analysis? • The regression procedure is used when you are interested in describing the linear relationship between the independent variables and a dependent variable. • A line in a two dimensional or two-variable space is defined by the equation Y=a+b*X • In full text: the Y variable can be expressed in terms of a constant (a) and a slope (b) times the X variable. • The constant is also referred to as the intercept, and the slope as the regression coefficient or B coefficient. • For example, GPA may best be predicted as 1+.02*IQ. Thus, knowing that a student has an IQ of 130 would lead us to predict that her GPA would be 3.6 (since, 1+.02*130=3.6).

What is the Regression Analysis? • In the multivariate case, when there is more than one independent variable, the regression line cannot be visualized in the two dimensional space, but can be computed just as easily. • For example, if in addition to IQ we had additional predictors of achievement (e.g., Motivation, Self- discipline) we could construct a linear equation containing all those variables. In general then, multiple regression procedures will estimate a linear equation of the form: • Y = a + b1*X1 + b2*X2 + ... + bp*Xp

1) Predicted and Residual Scores • The regression line expresses the best prediction of the dependent variable (Y), given the independent variables (X). • However, nature is rarely (if ever) perfectly predictable, and usually there is substantial variation of the observed points around the fitted regression line (as in the scatterplot shown earlier). The deviation of a particular point from the regression line (its predicted value) is called the residual value.

2) Residual Variance and R-square • The smaller the variability of the residual values around the regression line relative to the overall variability, the better is our prediction. • For example, if there is no relationship between the X and Y variables, then the ratio of the residual variability of the Y variable to the original variance is equal to 1.0. If X and Y are perfectly related then there is no residual variance and the ratio of variance would be 0.0. In most cases, the ratio would fall somewhere between these extremes, that is, between 0.0 and 1.0. • 1.0 minus this ratio is referred to as R-square or the coefficient of determination. This value is immediately interpretable in the following manner. If we have an R-square of 0.4 then we know that the variability of the Y values around the regression line is 1-0.4 times the original variance; in other words we have explained 40% of the original variability, and are left with 60% residual variability. • Ideally, we would like to explain most if not all of the original variability. The R-square value is an indicator of how well the model fits the data (e.g., an R-square close to 1.0 indicates that we have accounted for almost all of the variability with the variables specified in the model).

2) R-square • A mathematical term describing how much variation is being explained by the X. • R-square = SSR / SST • SSR – SS (Regression) • SST – SS (Total)

3) Adjusted R-square • Adjusted R-square is the adjusted value for R-square will be equal or smaller than the regular R-square. The adjusted R-square adjusts for a bias in R-square. • R-square tends to over estimate the variance accounted for compared to an estimate that would be obtained from the population. There are two reasons for the overestimate, a large number of predictors and a small sample size. So, with a small sample and with few predictors, adjusted R-square should be very similar the R-square value. Researchers and statisticians differ on whether to use the adjusted R-square. It is probably a good idea to look at it to see how much your R-square might be inflated, especially with a small sample and many predictors. • Adjusted R-square = 1 – [MSR / (SST/(n – 1))] • MSR – MS (Regression) • SST – SS (Total)

4) Coefficient R (Multiple R) • Customarily, the degree to which two or more predictors (independent or X variables) are related to the dependent (Y) variable is expressed in the correlation coefficient R, which is the square root of R-square. In multiple regression, R can assume values between 0 and 1. • To interpret the direction of the relationship between variables, one looks at the signs (plus or minus) of the regression or B coefficients. If a B coefficient is positive, then the relationship of this variable with the dependent variable is positive (e.g., the greater the IQ the better the grade point average); if the B coefficient is negative then the relationship is negative (e.g., the lower the class size the better the average test scores). Of course, if the B coefficient is equal to 0 then there is no relationship between the variables.

5) ANOVA • In general, the purpose of analysis of variance (ANOVA) is to test for significant differences between means. • At the heart of ANOVA is the fact that variances can be divided up, that is, partitioned. Remember that the variance is computed as the sum of squared deviations from the overall mean, divided by n-1 (sample size minus one). Thus, given a certain n, the variance is a function of the sums of (deviation) squares, or SS for short. Partitioning of variance works as follows. Consider the following data set:

6) Degree of Freedom (df) • Statisticians use the terms "degrees of freedom" to describe the number of values in the final calculation of a statistic that are free to vary. Consider, for example the statistic s-square.

7) S square & Sums of (deviation) squares • The statistic s square is a measure on a random sample that is used to estimate the variance of the population from which the sample is drawn. • Numerically, it is the sum of the squared deviations around the mean of a random sample divided by the sample size minus one. • Regardless of the size of the population, and regardless of the size of the random sample, it can be algebriacally shown that if we repeatedly took random samples of the same size from the same population and calculated the variance estimate on each sample, these values would cluster around the exact value of the population variance. In short, the statistic s squared is an unbiased estimate of the variance of the population from which a sample is drawn.

7) S square & Sums of (deviation) squares • When the regression model is used for prediction, the error (the amount of uncertainty that remains) is the variability about the regression line, . This is the Residual Sum of Squares (residual for left over). It is sometimes called the Error Sum of Squares. The Regression Sum of Squares is the difference between the Total Sum of Squares and the Residual Sum of Squares. Since the total sum of squares is the total amount of variability in the response and the residual sum of squares that still cannot be accounted for after the regression model is fitted, the regression sum of squares is the amount of variability in the response that is accounted for by the regression model.

8) Mean Square Error • ANOVA is a good example of why many statistical test represent ratios of explained to unexplained variability . It refers to an estimate of the population variance based on the variability among a given set of measures. It is an estimate of the population variance based on the average of all s-square within the several samples.

Agenda Overview of the Regression Statistical Significance Minitab

1) What is "statistical significance" (p-value) • The statistical significance of a result is the probability that the observed relationship (e.g., between variables) or a difference (e.g., between means) in a sample occurred by pure chance ("luck of the draw"), and that in the population from which the sample was drawn, no such relationship or differences exist. Using less technical terms, one could say that the statistical significance of a result tells us something about the degree to which the result is "true" (in the sense of being "representative of the population"). • More technically, the value of the p-value represents a decreasing index of the reliability of a result (see Brownlee, 1960). The higher the p-value, the less we can believe that the observed relation between variables in the sample is a reliable indicator of the relation between the respective variables in the population. • Specifically, the p-value represents the probability of error that is involved in accepting our observed result as valid, that is, as "representative of the population."

1) What is "statistical significance" (p-value) • For example, a p-value of .05 (i.e.,1/20) indicates that there is a 5% probability that the relation between the variables found in our sample is a "fluke." In other words, assuming that in the population there was no relation between those variables whatsoever, and we were repeating experiments like ours one after another, we could expect that approximately in every 20 replications of the experiment there would be one in which the relation between the variables in question would be equal or stronger than in ours. (Note that this is not the same as saying that, given that there IS a relationship between the variables, we can expect to replicate the results 5% of the time or 95% of the time; when there is a relationship between the variables in the population, the probability of replicating the study and finding that relationship is related to the statistical power of the design.). • In many areas of research, the p-value of .05 is customarily treated as a "border-line acceptable" error level. It identifies a significant trend. • f

2) What is "statistical significance" (F-test & t-test) • F test • The F test employs the statistic (F) to test various statistical hypotheses about the mean (or means) of the distributions from which a sample or a set of samples have been drawn. The t test is a special form of the F test. • F-value • F-value is the ratio of MSR/MSE. This shows the ratio of the average error that is explained by the regression to the average error that is still unexplained. Thus, the higher the F, the better the model, and the more confidence we have that the model that we derived from sample data actually applies to the whole population, and is not just an aberration found in the sample. • Significance of F • The value was computed by looking at standardized tables that consider the F-value and your sample size to make that determination. • If the significance of F is lower than an alpha of 0.05, the overall regression model is significant • t-test • The t test employs the statistic (t) to test a given statistical hypothesis about the mean of a population (or about the means of two populations).

Agenda Overview of the Regression Statistical Significance Minitab

Scatter Plot

X and Y Data Correlation

Building a Scatter Plot

Scatter Plot Example

Scatter Plot Input and Output

Simple Linear Regression • Regression analysis is a statistical technique • Used to model and investigate the relationship between two or more variables • The model is often used for prediction • Regression is a hypothesis test • Ha: The model is a significant predictor of the response. • May be used to analyze relationships between the “Xs,” or between “Y” and “X.” • Regression is a powerful tool, but can never replace engineering or manufacturing process knowledge about trends.

Regression Analysis Example

Regression Analysis Example - Output

Minitab RegressionCalculation Explanation of Output • The “p-values” for the constant (Y-intercept) and the predictor variables are read exactly as explained in Hypothesis Testing. • Ha: The factor is a significant predictor of the response. • The value s is the “standard error of the prediction” = sigma for individual data points. • R-square is the percent of variation explained by your model. • R-square (adjusted) is the percent of variation explained by your model, adjusted for the number of terms in your model and the number of data points. • The “p-value” for the regression is for whether the entire regression model is significant. • Ha: The model is a significant predictor of the response.

Confidence and Prediction Bands

Regression Analysis Input and Output

Regression Analysis – Confidence and Prediction Bands • A confidence band (or interval) • A measure of the certainty of the shape of the fitted regression line • In general, a 95% confidence band implies a 95% chance that the true line lies within the band. [Red lines] • A prediction band (or interval) • A measure of the certainty of the scatter of individual points about the regression line • In general, 95% of the individual points (of the population on which the regression line is based) will be contained in the band. [Blue lines]

Regression Analysis – Summary • Scatter Plots: Visual tool to establish a cause and effect relationship between the inputs and the outputs. • Simple Linear Regression • Statistical technique used to investigate the relationship between 2 variables • Ha: The factor is a significant predictor of the response • R2: percent of variation explained by yourmodel. In general, the closer R2 is to 1, the better the fit of the model • Prediction Intervals: 95% of data within the population falls within this band • Confidence Intervals: There exists a 95% chance that the true line of the population lies within the band • Prediction Interval: Can be used in statistical tolerancing

Hypothesis Tests – Summary

MGS 8020 Business Intelligence Analyze Nov 6, 2014