Problem to Project & Basic Statistical Analysis

Today's episode is all about Project Selection and Statistical Analysis, where we aim to provide you with practical guidance and insights. We understand that choosing the right projects can be overwhelming, but fear not! We are here to help you transform problems into well-defined projects that lead to growth and success. Building a strong pipeline of projects is essential for addressing organizational challenges and driving progress.

Statistical Analysis, though initially intimidating, is a fascinating realm that holds great power in continuous improvement. It enables us to summarise data, test hypotheses, and make evidence-based decisions that bring about impactful changes. In this episode, we will gently introduce you to three statistical analysis techniques: the '2 Sample T-Test', Correlation Analysis and Regression Analysis.

Consider this episode as a refreshing sip of a tropical cocktail of statistical knowledge, leaving you craving for more. And joining us today is Pranav Radhakrishnan, a distinguished Director at Novartis, a leading global pharmaceutical company. With his expertise in Operational Excellence and Lean Six Sigma, a Masters in Industrial Engineering, a pending MBA from Columbia University and a remarkable career journey, Pranav brings a unique and knowledgeable perspective to project selection and statistical analysis.

Together, we will explore practical strategies and inspiring experiences that will empower you to prioritize strategic projects, drive financial growth, and enhance engagement. Get ready to unlock the secrets of Project Selection and Statistical Analysis with Pranav's invaluable insights.

Key Takeaways:

Learn to select the right tool for the right problem. It comes with experience.  Pranav gave a great method of taking problems to projects, that directly relate to the business objectives.  Get people together, directors, department leads, supervisors, front line employees. All roles – get them in a room. Look at overarching operational strategic objectives. Large scale objectives. 3 or 4 is sufficient. Then rank these by importance with a score of 1-10. 10 being most important. As a group, brainstorm using post it notes to document the biggest opportunities and issues that people see. No rules, just get them all noted down. If struggling, try a negative brainstorm and then flip the outputs to create opportunities. Then group the post it notes by categories e.g. processes or services.  Think about the effort vs ease vs impact of the issues noted. An impact matrix is a great tool for this. Colour code them if it helps.  Next up, use a cause and effect matrix style process to help determine which to focus on. Rank each problem by importance, on a scale of 0-10 (10 being critical). Then rank each opportunity one by one to see correlation to each strategic goal. Give it a 0 if no relation, a 1 if minimal, a 5 if some correlation and 9 for highly correlated. Do this for each issue. This will generate more debate and discussion. This is good. Come to a consensus to Identify which issues relate to each of the pre-defined objectives. Facilitate the conversation, staying neutral if you can. Now for each problem, simply add the scores up. You’ll be left with a valuable list of issues, ranked and aligned to your organisation’s strategic direction. Super useful mechanism to build an impactful roadmap for both short term and longer-term transformation.  Statistical analysis brings objectivity, rigor, and evidence-based decision-making to CI. It helps uncover insights, quantify uncertainties, test hypotheses, predict future outcomes, and evaluate performance, enabling organizations to stay competitive in a rapidly evolving business environment. Today we focus on 3 commonly used statistical analysis approaches: We looked at the 2 sample T test technique This is used to compare the means of two independent groups or samples to determine if they are significantly different from each other. It helps to assess whether any observed differences between the groups are statistically significant or simply due to random chance. Let's say a coffee shop owner wants to determine whether there is a significant difference in the average time it takes to serve customers during the morning shift and the afternoon shift. The owner collects data on the time taken to serve 50 customers during the morning shift and 50 customers during the afternoon shift. The owner wants to know if there is evidence to suggest that one shift is significantly faster than the other. By conducting the two-sample t-test, the coffee shop owner can determine if there is a statistically significant difference in serving times between the morning and afternoon shifts. This information can be valuable for making staffing decisions, optimizing workflow, or identifying areas for improvement in service efficiency. We explored Correlation analysis. This technique is used to measure the strength and direction of the relationship between two variables. It helps to determine whether there is a systematic association between the variables and to what extent they vary together. Consider a researcher who wants to examine the relationship between studying hours and exam scores among a group of students. The researcher collects data on the number of hours each student spends studying for an exam and their corresponding exam scores. The researcher wants to determine if there is a relationship between the two variables. By conducting correlation analysis, the researcher can gain insights into the relationship between studying hours and exam scores. This information can be useful for students, educators, and policymakers in understanding the importance of study habits and their impact on academic performance. Then finally we took a shallow dip into Regression analysis. This technique is used to model and analyze the relationship between a dependent variable and one or more independent variables. It helps to understand how changes in the independent variables are associated with changes in the dependent variable and to make predictions or estimate values based on the observed data. Suppose a real estate agent wants to determine the relationship between the size of a house (in square feet) and its selling price. The agent collects data on recently sold houses, recording the size of each house and its corresponding selling price. The agent wants to predict the selling price of a house based on its size. By conducting regression analysis, the real estate agent can estimate the relationship between house size and selling price. This information can be useful for pricing properties, advising clients, and making informed decisions in the real estate market.

Please don't forget to show your support and help the Ever-So-Lean Podcast thrive by taking a moment to rate and review it, your feedback matters!

 

Contact Pranav via his LinkedIn profile

 

Find out more about the British Quality Foundation at www.bqf.org.uk

To find out more about the certifications available with Catalyst Consulting Ltd, visit www.catalystconsulting.co.uk

 

Enrol in the Online Improvement Kata & Coaching Kata Basics Course. The only Kata training created by Jeffrey K. Liker and Mike Rother!

 

Find out more about the Ever-So-Lean Podcast at www.eversolean.com