Workforce Scheduling in Logistics Hubs Dataset

This Workforce Scheduling and Operational Efficiency Dataset captures real-time, high-frequency operational data collected from a logistics hub over a 7-year period spanning from January 1, 2018, to December 31, 2024. The dataset includes hourly records for a diverse set of workforce, environmental, and operational features, reflecting actual dynamics observed in large-scale logistics operations.

The dataset is structured to support analysis of workforce management, shift optimization, employee fatigue assessment, overtime trends, and task sufficiency. It provides a detailed view of how human factors, environmental conditions, and task complexity interact in high-demand logistics environments.

Feature Overview

A. Temporal and Identification

DateTime: Timestamp of observation at hourly intervals.

Employee_ID: Unique identifier for individual workforce members.

B. Shift and Job-Related Attributes

Shift_Timing: Operational shift per hour (Morning, Evening, Night).

Role_Position: Assigned job function (Loader, Supervisor, Driver).

Skill_Level: Skill classification of the employee (Beginner, Intermediate, Advanced).

Shift_Preference: Preferred working time of the employee.

C. Experience and Workforce Behavior

Work_Experience_Years: Years of experience in logistics or similar roles.

Availability_Hours: Number of hours the worker is available for scheduling.

Previous_Shift_Fatigue: Fatigue index based on preceding work hours.

Training_Hours_Completed: Cumulative training hours completed by each employee.

Absenteeism_Rate: Relative absenteeism level based on historical behavior.

D. Operational Load and Environment

Daily_Shipment_Volume: Shipment volume handled per shift.

Shipment_Type: Classification of the shipment (Fragile, Perishable, Standard).

Demand_Fluctuations: Measure of variation in demand at the hour level.

Hub_Size: Relative size and capacity of the logistics hub.

Temperature and Humidity: Environmental readings at the logistics location.

E. Scheduling Constraints

Required_Labor_Units: Hourly workforce requirement estimate.

Historical_Delay_Data: Frequency of past delays associated with tasks.

Max_Shift_Duration: Maximum allowable work hours per shift.

Min_Rest_Period: Minimum rest period required between consecutive shifts.

Overtime_Limits: Permissible overtime hours per worker.

Shift_Overlap: Measure of overlap between shift transitions.

F. Performance and Feedback Metrics

On_Time_Completion_Rate: Task completion punctuality.

Worker_Efficiency_Score: Productivity score based on output vs. time.

Schedule_Adherence: Alignment with pre-assigned shift schedules.

Employee_Satisfaction_Rating: Score derived from HR or operational evaluations.

Target Labels for Prediction Tasks

The dataset includes multi-label classification targets to enable the development of supervised models:

Optimal_Shift_Assignment: Binary outcome (Assigned / Not Assigned) for schedule feasibility.

Overtime_Prediction: Indicator for whether overtime was required (Yes / No).

Employee_Fatigue_Risk: Categorical fatigue risk (High, Medium, Low).

Shift_Coverage_Sufficiency: Binary label indicating if staffing was sufficient (Sufficient / Insufficient).

Highlights

Time Span: 7 years (2018–2024), hourly resolution.

Multi-Domain Variables: Covers human, environmental, operational, and scheduling data.

Multi-Label Supervision: Enables simultaneous prediction of multiple labor-related outcomes.

Real-Time Nature: Suitable for online analytics, edge computing, and federated learning scenarios.

Use Cases: Workforce optimization, logistics planning, fatigue risk monitoring, shift planning, ML benchmarking.