The Employee Attrition Control project aims to help a company analyze and address employee turnover. By leveraging data on existing and former employees, this project uses analytics techniques to gain insights into the factors contributing to attrition and make recommendations for mitigating it.
This project seeks to answer the following questions:
- What type of employees are leaving?
- Which employees are prone to leave next?
- What recommendations can be made to control attrition?
To address these questions, the project employs the following analytics techniques:
- Analyzing Employee Characteristics:
- Compare satisfaction levels, evaluation scores, number of projects, average monthly hours, time spent at the company, work accident history, promotion history, department distribution, and salary levels between existing and former employees.
- Identify patterns and trends to determine the characteristics of employees who are leaving the company.
- Predictive Modeling:
- Build a machine learning model using the existing employee dataset as training data and the attrition status as the target variable.
- Use the model to predict the probability of attrition for each employee in the existing employee dataset.
- Identify employees who are more likely to leave next based on the predicted probabilities.
Based on the analysis and predictions, the project offers the following recommendations to control employee attrition:
- Improve job satisfaction by addressing factors leading to low satisfaction levels. Conduct employee surveys, provide career development opportunities, and foster a positive work environment.
- Enhance employee engagement through recognition and rewards, regular feedback sessions, and opportunities for skill enhancement.
- Promote work-life balance by monitoring workload and offering flexible scheduling and employee wellness programs.
- Provide clear career growth opportunities and recognize employees' contributions through promotions.
- Analyze departments with higher attrition rates and address any issues related to management, workload distribution, or work environment specific to those departments.
- Regularly review and benchmark salary levels to ensure competitiveness. Provide attractive benefits packages and consider performance-based incentives.
- Offer resources for personal and professional development, mentorship programs, and avenues for open communication to support employees.
To use this project, follow these steps:
- Ensure the availability of the "Existing employees" and "Employees who have left" datasets.
- Execute the analytics techniques described in the project, such as data analysis, predictive modeling and generating insights.
- Apply the recommendations provided to control employee attrition based on the project's findings.
The project requires the following dependencies:
- Python 3.x
- Required Python libraries (e.g., pandas, NumPy, scikit-learn, seaborn, matplotlib)
This project was developed by Nafisa Lawal Idris. Contributions and feedback are welcome.
This project is licensed under the MIT License.