Conceptualized by Ankit Srivastava, SlideScope
Modern manufacturing environments generate data continuously through PLCs, sensors, machines, and production systems. Learning how to convert this raw operational data into useful business insights is an important skill for students entering data analytics, business intelligence, and industrial analytics.
The Multi-PLC Production Dataset, conceptualized by Ankit Srivastava of SlideScope, is designed as a practical training dataset for this purpose. It contains production information generated from four PLCs for 08 October 2026, with hourly observations throughout the day.
Each PLC has 24 hourly records, resulting in a total of 96 records across the four CSV files. Instead of providing students with one consolidated file, the dataset deliberately separates the information into four files:
PLC_01_08_Oct_2026.csvPLC_02_08_Oct_2026.csvPLC_03_08_Oct_2026.csvPLC_04_08_Oct_2026.csv
Get the dataset Files Here: https://www.kaggle.com/datasets/slidescope/multi-plc-industrial-production-dataset
This structure creates a realistic learning environment because manufacturing organizations frequently receive operational data from different machines, production lines, PLCs, or systems.
The dataset contains several useful production fields, including Date, Time, PLC ID, Production Quantity, Good Quantity, Reject Quantity, Target Quantity, Downtime Minutes, Cycle Time, Temperature, Pressure, Machine Status, and Shift.
Students can therefore investigate both production performance and operational conditions.
For example, they can compare the production output of different PLCs, identify machines with higher rejection rates, examine downtime, compare actual production against targets, and investigate whether temperature or pressure variations appear alongside production changes.
The Shift field also allows students to perform Morning, Evening, and Night shift analysis.
Another important feature of the dataset is that the four PLCs do not behave identically. This allows students to discover differences rather than simply producing identical charts for every machine.
The dataset is therefore suitable for teaching the complete analytics workflow:
Multiple CSV Files → Data Cleaning → Data Transformation → Data Modeling → KPI Creation → Visualization → Business Insights
It can be used by beginners learning Power BI, Tableau, or Python, while also providing enough structure to introduce concepts such as calculated measures, filtering, grouping, time-based analysis, and machine-level comparisons.
How to Use the Dataset in Power BI
Power BI is particularly useful for this dataset because students can learn how to combine multiple files using Power Query.
Instead of importing the four CSV files individually, students can place them inside a single folder and use the Folder connector in Power BI.
Go to:
Home → Get Data → Folder
Select the folder containing the four PLC files. Power BI can then combine the files into a single table.
Students should inspect the automatically generated query and ensure that fields such as Date and Time have the correct data types. They can also create a proper Date/Time column by combining the Date and Time fields.
The resulting dataset can then be used to create DAX measures such as:
Total Production
Total Production = SUM(Production_Qty)
Total Good Production
Good Production = SUM(Good_Qty)
Total Rejects
Total Rejects = SUM(Reject_Qty)
Reject Rate
Reject Rate = DIVIDE([Total Rejects], [Total Production])
Target Achievement
Target Achievement = DIVIDE([Total Production], SUM(Target_Qty))
Students can create KPI cards for production, good units, rejects, downtime, and target achievement.
Useful visuals include a PLC comparison chart, hourly production line chart, Good vs Reject chart, downtime analysis, production by shift, and Target vs Actual production.
The final dashboard can include slicers for PLC, Shift, Machine Status, and Time, allowing users to interactively investigate production performance.
The key lesson is not only creating a dashboard, but learning how Power BI can turn multiple operational files into a unified analytical model.
How to Use the Dataset in Tableau
Tableau can also be used to analyze the four PLC files and create an interactive manufacturing dashboard.
Students can begin by connecting Tableau to one of the CSV files. If the four files have the same structure, they can bring the files together using Tableau’s file and union capabilities.
The objective is to create one analytical table containing all four PLCs while retaining the PLC_ID field so that machine-level comparisons remain possible.
After connecting the data, students should verify the data types. Date and Time can be converted into appropriate temporal fields, while production, target, downtime, temperature, pressure, and cycle-time fields should be treated as numerical measures.
Students can then create calculated fields such as:
Reject Rate
SUM([Reject_Qty]) / SUM([Production_Qty])
and:
Target Achievement
SUM([Production_Qty]) / SUM([Target_Qty])
A Tableau dashboard can then be constructed using multiple worksheets.
For example, one worksheet can show Total Production by PLC, another can show Hourly Production, and another can compare Good Quantity and Reject Quantity.
A downtime visualization can identify which PLC experienced the greatest amount of downtime.
Students can also create a Shift Analysis comparing Morning, Evening, and Night production.
The dashboard can contain filters for PLC_ID, Shift, Machine Status, and Date.
A useful training exercise is to ask students to identify the best-performing PLC based on production volume, reject rate, downtime, and target achievement rather than relying on a single KPI.
This teaches an important analytical concept: the machine producing the most units is not necessarily the most efficient machine.
How to Use the Dataset in Python
Python provides students with an opportunity to perform the same production analysis programmatically.
The first step is to place all four CSV files in the same directory. Using pandas, students can read each file and combine them into one DataFrame.
For example:
import pandas as pd
import glob
files = glob.glob("PLC_Production_Data/*.csv")
dfs = [pd.read_csv(file) for file in files]
production = pd.concat(dfs, ignore_index=True)Students can then inspect the dataset using:
print(production.head())
print(production.info())
print(production.describe())The PLC-level production can be calculated using:
production.groupby("PLC_ID")["Production_Qty"].sum()Reject rates can also be calculated:
production["Reject_Rate"] = (
production["Reject_Qty"] /
production["Production_Qty"]
) * 100Students can use groupby() to compare production by PLC and shift:
production.groupby(
["PLC_ID", "Shift"]
)["Production_Qty"].sum()For visualization, students can use Matplotlib to create hourly production charts, PLC comparison charts, and reject-rate visualizations.
Python can also be used to investigate relationships between operational variables. For example, students could examine whether higher temperature or pressure values are associated with changes in production or rejects.
The project can therefore progress from basic pandas operations to exploratory data analysis, visualization, correlation analysis, and eventually machine-learning exercises.
The most important learning outcome is understanding the complete workflow:
Read multiple files → Combine data → Clean data → Calculate KPIs → Analyze patterns → Visualize insights → Communicate findings.
This makes the dataset a practical bridge between classroom exercises and real-world industrial data analytics.
Conclusion
The Multi-PLC Production Dataset provides students with a compact but realistic manufacturing analytics environment. With four PLCs, four separate CSV files, 24 hourly records per PLC, and 96 total observations, it demonstrates how operational data can be transformed into meaningful analytical information.
The same dataset can be approached through Power BI for business intelligence, Tableau for interactive visualization, and Python for programmable data analysis.
Conceptualized by Ankit Srivastava of SlideScope, the dataset is particularly useful for teaching students an important real-world skill: data rarely arrives perfectly prepared in a single file. Analysts must combine, clean, transform, analyze, and communicate information before meaningful business decisions can be made.