Trend & Seasonality Analysis in Minitab – Tabtrainer® Tools

Time series analysis with Minitab including inroduction to the Minitab's user interface

In this session, participants gain a comprehensive understanding of Minitab's user interface and its key functionalities befor start with time Series Analysis. They learn how to navigate, configure, and manage datasets for time series analysis with Minitab efficiently. Key takeaways include:

1. User Interface Overview:

   • Structure of Minitab's interface, including worksheets, the output window, and navigation pane.

   • How to enlarge, reduce, or customize different interface sections for better usability.

2. Basic Operations:

   • Importing datasets from various sources, such as Excel, databases, and Minitab projects.

   • Setting up default file locations to streamline workflows.

3. Data Management:

   • Overview of worksheets, including how to add, edit, and describe data columns.

   • Differentiating between text, numeric, and date data types.

4. Graphical and Statistical Tools:

   • Using Minitab's graphical tools, like scatter plots and matrix plots, to identify patterns and trends.

   • Understanding statistical analyses, such as ANOVA, regression, and descriptive statistics.

5. Automation with Macros:

   • Utilizing the History window to generate and edit macros for recurring calculations.

   • Automating workflows with conditional programming in the Command Line.

6. Documentation:

   • Adding comments and notes to columns, worksheets, and projects for enhanced clarity.

   • Exporting key insights and findings for presentations and reporting.

By the end of this session, participants will be proficient in navigating Minitab and preparing datasets for time Series Analysis.

In this chapter, participants will learn:

1. Purpose of Analysis:

   • Understand the importance of evaluating seasonal trends and patterns in production scrap rates as part of a quality improvement project.

2. Data Preparation:

   • Identify relevant data for analysis, excluding unreliable records (e.g., data prior to January 1, 2017, due to inconsistency in error measurement).

3. Trend Analysis:

   • Use time series plots to explore trends and patterns in scrap rates, focusing on specific time periods (e.g., months, years).

4. Contextual Influences:

   • Assess the impact of seasonal workforce changes (e.g., auxiliary staff during summer and Christmas holidays) on production quality.

5. Key Metrics:

   • Identify fiscal years with the highest and lowest scrap rates, helping pinpoint opportunities for improvement.

This exercise emphasizes how data preparation and proper analysis techniques can uncover valuable insights for decision-making and quality control.

This chapter explains the data import and preparation process in Minitab using an Excel dataset as an example. Key takeaways include:

1. Data Import Capabilities:

   • Learn how to import datasets from non-Minitab file formats like Excel or text files into Minitab.

   • Understand Minitab’s worksheet size limitations (maximum 4000 columns and 10 million rows) and practical IT constraints (150 million filled cells).

2. Data Conversion:

   • Explore how Minitab automatically converts Excel files into worksheets without modifying the original file.

   • Verify the accuracy of data conversion using Minitab's preview tools.

3. Naming and Saving Projects:

   • Assign meaningful names to worksheets and projects for better organization.

   • Save projects in predefined directories to ensure consistency during training sessions.

4. Dataset Overview:

   • Analyze the structure of imported datasets, including column formats (e.g., date and numeric values).

   • Use worksheet information tools to gain insights, such as the total number of rows and the time span covered by the data.

5. Data Visualization:

   • Create a time series plot to obtain a comprehensive overview of the dataset and identify potential trends or patterns.

This chapter equips learners with foundational skills for importing, organizing, and visualizing data in Minitab efficiently.

This chapter introduces the process of creating a time series plot in Minitab to visualize data trends and patterns over time. Key points include:

1. Purpose of Time Series Plots:

   • Time series plots are used to depict data in chronological order, allowing for a clear visual representation of trends, such as the scrap quantities over the last 20 years.

2. Types of Time Series Plots:

   • Simple: Displays one variable in chronological order (used in this example).

   • With Groups: Shows values grouped by categories.

   • Multiple: Depicts multiple data series.

   • Multiple with Groups: Combines multiple data series with categorical grouping.

   The Simple type was selected for this case, as the dataset contains one column (C2) with numerical scrap rates and one (C1) with dates.

3. Creating the Plot:

   • Data from column C2 (scrap rates) was used to generate the plot.

   • The x-axis represents production dates, while the y-axis shows daily scrap quantities.

4. Enhancing the Visualization:

   • The x-axis was updated to display unambiguous dates instead of sequential row numbers, improving clarity.

5. Initial Observations:

   • The plot reveals a general increase in scrap quantities over time.

   • However, further trends—such as variations across years or weekdays—are not immediately clear.

6. Next Steps:

   • The worksheet will be refined to focus on valid data as defined by quality assurance restrictions, enabling more detailed analysis in subsequent steps.

This chapter emphasizes the importance of selecting appropriate visualization methods and iteratively refining data for deeper insights.

This chapter focuses on data extraction and preparation to ensure accurate and meaningful time series analysis. Key points include:

1. Data Extraction:

   • Only data from January 1, 2017 onward is included in the analysis, as earlier records are deemed unreliable.

   • Extracting specific data subsets involves creating a new worksheet using Minitab’s "Subset Worksheet" function, ensuring the original dataset remains unchanged.

2. Steps for Data Extraction:

   • Define conditions for inclusion (e.g., "date is after January 1, 2017").

   • Create a new worksheet titled "Scrap rate from January 1, 2017" containing only the relevant data.

   • Verify that no missing values are present before proceeding.

3. Creating a New Time Series Plot:

   • Ensure the correct worksheet is active before generating graphs or conducting analyses.

   • Follow the steps to create a Simple Time Series Plot and enhance the x-axis by adding explicit dates using the "Stamp" option.

4. Identifying Limitations:

   • The new time series plot shows clearer trends but lacks sufficient detail to evaluate variations, such as differences across weekdays.

5. Adding Weekday Information:

   • To better understand weekday-based variations, the chapter introduces the need to extract and add a new column listing weekdays (Monday to Sunday) in text form. This will facilitate a more detailed analysis in subsequent steps.

This chapter emphasizes the importance of refining datasets, iteratively improving visualizations, and adding contextual information to uncover deeper insights into data patterns.

This chapter focuses on further refining the dataset and analyzing potential outliers to enhance the clarity and accuracy of time series plots. Key points include:

1. Adding Weekday Information:

   • A new column, "Weekdays", is created to list the associated weekday (e.g., Monday to Sunday) in text format. This provides additional context for analyzing weekday-specific trends.

2. Excluding Non-Working Days:

   • Saturdays and Sundays are excluded from the dataset to focus solely on production days (Monday to Friday).

   • A new worksheet is created: "Scrap rate from January 1st, 2017, no weekends".

3. Updated Time Series Plot:

   • A revised time series plot is generated using the refined dataset, showing only working days starting from January 2, 2017.

   • The plot reveals patterns such as:

     • Zero scrap quantities on certain days.

     • Periods with exceptionally high scrap rates.

4. Analyzing Zero Scrap Values:

   • Data points with zero scrap quantities are identified and analyzed using Minitab’s "Brush" tool and "Set Identification Variables" function.

   • Marked points are associated with holidays (e.g., New Year’s Day, Labor Day, Christmas), explaining the absence of production and scrap quantities on those days.

5. Significance of Identification Variables:

   • Identification variables are used to quickly gather additional context for outlier data points. This helps determine whether these points represent significant trends or should be excluded as anomalies.

This chapter demonstrates how to refine datasets by excluding irrelevant data and leveraging Minitab tools to investigate anomalies, providing a clearer and more accurate basis for time series analysis.

This chapter focuses on eliminating holidays from the dataset and interpreting seasonal fluctuations in scrap rates. Key takeaways include:

1. Removing Holidays from the Dataset:

   • Holidays identified in the previous step are excluded to refine the dataset further.

   • Using Minitab’s "Brush" tool, specific data points (holidays) are marked and excluded directly via the "Subset Worksheet" function.

   • A new worksheet, "Scrap rate from January 1st 2017, no Weekends, no Holidays", is created to focus only on production days.

2. Creating the Final Time Series Plot:

   • A new time series plot is generated using the refined dataset, ensuring that holidays and non-working days are excluded.

   • The x-axis (dates) and y-axis (scrap rates) are updated as before to present a clean, focused visualization.

3. Interpreting Seasonal Fluctuations:

   • The plot reveals higher scrap quantities during summer vacations and around Christmas compared to other months.

   • This seasonal variation is attributed to temporary workers employed during these periods due to regular staff shortages, who may lack the technical qualifications of permanent employees.

This chapter highlights the importance of iterative data refinement and demonstrates how to identify and interpret seasonal trends using well-prepared time series plots.

This chapter delves into the reasons for high scrap rates during specific periods and concludes with actionable recommendations. Key takeaways include:

1. Analyzing High Scrap Rate Periods:

   • By using the Brush tool in edit mode, specific high-value data clusters (data clouds) are identified for further investigation.

   • Data from summer vacations in 2017, 2018, and 2019 reveals a consistent pattern: significantly higher scrap rates occur exclusively on Mondays and Fridays during these periods.

2. Seasonal Insights:

   • Higher scrap quantities during summer vacations correlate with the employment of temporary staff, who may lack sufficient technical qualifications.

   • These patterns exacerbate the already lower production performance typically seen during vacation periods.

3. Other Observations:

   • A notable decrease in scrap rates is observed in the first quarter of 2019 compared to previous years, indicating potential process improvements or shifts in production quality.

4. Final Recommendations:

   • Conduct a Quality Audit: Assess whether vacation planning requires optimization to better distribute workloads and reduce scrap rates.

   • Train Temporary Staff: Introduce targeted training programs for temporary employees to enhance their qualifications and reduce errors.

5. Importance of Clean Data Preparation:

   • The clear identification of trends and tendencies was only possible due to rigorous data preparation, including the exclusion of irrelevant data points (weekends and holidays).

This chapter emphasizes the value of data-driven decision-making and highlights how well-prepared datasets can uncover actionable insights for quality and operational improvements.

This chapter underscores the significance of data cleaning and demonstrates how to document and present the results effectively. Key points include:

1. Importance of Data Cleaning:

   • Comparing the initial time series plot with the final cleaned plot highlights the impact of thorough data preparation.

   • By extracting relevant information, adding contextual details (e.g., weekdays), and removing non-significant data (e.g., weekends and holidays), chronological trends became clearer and more interpretable.

2. Exporting Results to Microsoft PowerPoint:

   • Time Series Plot:

     • The final time series plot can be directly exported to Microsoft PowerPoint using the dropdown menu in Minitab.

     • If PowerPoint is closed, it will open automatically; if open, the plot is appended as the last slide.

   • Worksheet Information:

     • The summary of the final worksheet (e.g., containing 750 rows of data) can also be exported to PowerPoint as a new slide.

3. Final Documentation:

   • Combining visualizations (time series plot) and metadata (worksheet information) in PowerPoint ensures a comprehensive presentation of findings, useful for reporting and decision-making.

This chapter highlights the critical role of data cleaning in producing meaningful insights and provides a seamless method to document and communicate results effectively.

This training session focused on data management and analysis of large, uncleaned datasets, using time series plots as the primary visualization tool. Here are the key takeaways:

1. Introduction to Time Series Plots:

   • Time series plots were used to chronologically visualize scrap rate data spanning approximately 20 years of production at Smartboard Company.

   • The initial plot, based on raw data, revealed limited insights due to high data density and lack of refinement.

2. Data Cleaning and Extraction:

   • The session emphasized the importance of data cleansing for uncovering meaningful trends.

   • Using the subset worksheet function, non-relevant data such as weekends and holidays was excluded step by step from the dataset.

   • Additional information, such as weekdays, was incorporated to provide more context and enable deeper analysis.

3. Interpreting Refined Data:

   • After refining the dataset, clear trends and tendencies in scrap rates were identified:

     • Periodic spikes in scrap rates during specific times, such as Mondays, Fridays, summer vacations, and Christmas.

     • A noticeable increase in overall scrap rates after a certain point in time.

   • These insights allowed Smartboard Company’s management to address their specific questions about scrap rate trends across production days, months, and years.

4. Finalization and Documentation:

   • A refined time series plot, based on only relevant data, provided clear and actionable insights.

   • The analysis was saved under the project name "Time Series Plot" to preserve results and ensure no data loss.

This session highlighted the critical role of data preparation, refinement, and visualization in answering key business questions and provided participants with the tools to effectively manage and analyze large datasets.