Skip to content
0

Predicting Industrial Machine Downtime: Level 1

📖 Background

You work for a manufacturer of high-precision metal components used in aerospace, automotives, and medical device applications. Your company operates three different machines on its shop floor that produce different sized components, so minimizing the downtime of these machines is vital for meeting production deadlines.

Your team wants to use a data-driven approach to predicting machine downtime, so proactive maintenance can be planned rather than being reactive to machine failure. To support this, your company has been collecting operational data for over a year and whether each machine was down at those times.

In this first level, you're going to explore and describe the data. This level is aimed towards beginners. If you want to challenge yourself a bit more, check out level two!

💾 The data

The company has stored the machine operating data in a single table, available in 'data/machine_downtime.csv'.

Each row in the table represents the operational data for a single machine on a given day:
  • "Date" - the date the reading was taken on.
  • "Machine_ID" - the unique identifier of the machine being read.
  • "Assembly_Line_No" - the unique identifier of the assembly line the machine is located on.
  • "Hydraulic_Pressure(bar)", "Coolant_Pressure(bar)", and "Air_System_Pressure(bar)" - pressure measurements at different points in the machine.
  • "Coolant_Temperature", "Hydraulic_Oil_Temperature", and "Spindle_Bearing_Temperature" - temperature measurements (in Celsius) at different points in the machine.
  • "Spindle_Vibration", "Tool_Vibration", and "Spindle_Speed(RPM)" - vibration (measured in micrometers) and rotational speed measurements for the spindle and tool.
  • "Voltage(volts)" - the voltage supplied to the machine.
  • "Torque(Nm)" - the torque being generated by the machine.
  • "Cutting(KN)" - the cutting force of the tool.
  • "Downtime" - an indicator of whether the machine was down or not on the given day.

Machine Downtime Analysis Report

Executive Summary

This report presents the findings from the analysis of machine downtime data collected from the shop floor. By examining operational parameters such as pressure, temperature, vibration, and torque, the goal is to identify potential predictors for machine failure. The analysis includes an overview of the dataset, a review of key metrics, and actionable insights to help improve machine uptime and support proactive maintenance strategies.

Key findings from the analysis:

  1. The first reading of machine downtime was recorded on November 24, 2021, and the last reading was recorded on July 3, 2022.
  2. The average torque during the reporting period was found to be 25.23 Nm.
  3. The assembly line with the highest readings of machine downtime is Shopfloor-L1.

1. What is the first and last date readings were taken on?

The dataset spans a period from November 24, 2021, to July 3, 2022. These dates mark the range of machine downtime readings available for analysis.

  • First reading date: November 24, 2021
  • Last reading date: July 3, 2022

Understanding this range is crucial for identifying trends in machine downtime over time, helping us better align maintenance schedules.

2. What is the average Torque?

The average torque recorded during the analysis period was 25.23 Nm. Torque is an important indicator of machine performance, and this metric helps establish baseline operating conditions. Monitoring changes in torque values over time can provide valuable insights into the machine’s health, indicating potential problems when values deviate significantly from this average.

3. Which assembly line has the highest readings of machine downtime?

The assembly line with the highest readings of machine downtime is Shopfloor-L1. By isolating rows where 'Machine_Failure' occurred, I identified that this assembly line experienced more downtime than others. This insight is crucial for allocating resources for targeted maintenance and understanding which areas of production need improvement.

Actionable Recommendation:

  • Targeted Maintenance for Shopfloor-L1: Given the high levels of downtime on Shopfloor-L1, I recommend prioritizing maintenance and further investigation into the root causes of failure. Regular monitoring of relevant parameters (such as temperature, pressure, vibration, and torque) should be implemented to predict potential failures before they occur.

Visualizations and Analysis

The visualizations below provide a deeper understanding of the data trends and distributions:

  1. Machine Downtime Over Time: A line plot showing the number of machine failures per day over the period. This visual can highlight periods of frequent breakdowns, which may suggest areas requiring more focused attention.

  2. Downtime by Assembly Line: A bar chart visualizing the total number of downtime occurrences across each assembly line, confirming that Shopfloor-L1 has the highest downtime.

  3. Torque Distribution: A histogram illustrating the distribution of torque values during the period, helping to understand how torque values correlate with machine performance and downtime.

Conclusion

In conclusion, this analysis provides a strong foundation for future predictive maintenance efforts. By focusing on Shopfloor-L1, where machine downtime is most prevalent, and closely monitoring parameters like torque, temperature, and vibration, the company can implement more effective maintenance strategies, reduce downtime, and improve overall operational efficiency.

Appendix:

  • Code and Methodology: The analysis was performed using Python, with libraries such as pandas, matplotlib, and seaborn for data manipulation and visualization. Key data processing steps included filtering downtime events and aggregating data by relevant variables (e.g., Date, Assembly_Line_No).
import pandas as pd
downtime = pd.read_csv('data/machine_downtime.csv')
downtime.head()

1. What is the first and last date readings were taken on?

downtime['Date'] = pd.to_datetime(downtime['Date'], errors='coerce')

first_reading = downtime['Date'].min()
last_reading = downtime['Date'].max()

print(f'First reading was taken on {first_reading}')
print(f'Last reading was taken on {last_reading}')

2. What is the average Torque?

avg_torque = downtime['Torque(Nm)'].mean()

print(f'Average Torque is {avg_torque:.2f}')

3. Which assembly line has the highest readings of machine downtime?

assembly_line_downtime = downtime[downtime['Downtime'] == 'Machine_Failure']

if not assembly_line_downtime.empty:
    
    highest_downtime_assembly_line = assembly_line_downtime['Assembly_Line_No'].value_counts().idxmax()
else:
    highest_downtime_assembly_line = None

print(f'{highest_downtime_assembly_line} has the highest readings of machine downtime.')

Visualisations

  1. Downtime over Time (Line Plot)
import matplotlib.pyplot as plt
import seaborn as sns

sns.set(style="whitegrid")

plt.figure(figsize=(10, 6))
downtime_over_time = downtime.groupby('Date')['Downtime'].apply(lambda x: (x == 'Machine_Failure').sum())
downtime_over_time.plot(kind='line', color='red', lw=2)
plt.title('Machine Downtime Over Time')
plt.xlabel('Date')
plt.ylabel('Downtime Occurrences')
plt.xticks(rotation=45)
plt.tight_layout()
plt.show()
  1. Downtime by Assembly Line (Bar Chart)
plt.figure(figsize=(10, 6))
sns.countplot(data=downtime, x='Downtime')
plt.title('Total Downtime by Assembly Line')
plt.xlabel('Total Downtime')
plt.ylabel('Assembly Line')
plt.tight_layout()
plt.show()
‌
‌
‌