A Survey of Machine Learning Use Cases (Applications) in Project Planning and Scheduling Process

Journal of Artificial Intelligence and Big Data | Vol 3, Issue 1

Table 1. Summary of Recent Studies on ProjectScheduling, Planning, and Optimization Approaches

Reference	Study on	Approach	Key Findings	Challenges / Limitations	Future Directions
Wang et al. (2019)	Proactive and reactive multi-project scheduling	An approach for genetic simulated annealing that incorporates proactive scheduling into reactive scheduling and makes use of buffer change and crossover operators	Generates alternative schedules efficiently and protects optimal solutions in early algorithm stages	Computational complexity and dependency on algorithm parameter tuning	Enhancement of algorithm scalability and application to real-time dynamic project environments
Shimoda, Wilairath & Kounosu (2019)	Work Breakdown Structure (WBS) creation	Two-layer activity structure combining PMBOK process groups and V-model steps	Improves beginner understanding and reduces scope loss in WBS creation	Validation mainly limited to student-based experiments	Application in industrial-scale projects and empirical validation with experienced project managers
Zachko & Kobylkin (2018)	Evacuation efficiency in mass-occupancy facilities	Simulation modeling using AnyLogic multi-agent environment and event-based evacuation scenarios	Enables visualization and efficiency analysis of evacuation processes	Focused on specific evacuation scenarios and infrastructure types	Extension to broader emergency scenarios and integration with real-time sensor data
Kusturica et al. (2018)	Project task duration estimation	Modular prototype with plug-ins based on state-of-the-art planning processes	Demonstrates potential benefits for improving project duration estimation accuracy	Prototype-level implementation with limited empirical evaluation	Full-scale deployment and validation across diverse project domains
Hameed et al. (2017)	Staffing and scheduling optimization	Particle Swarm Optimization (PSO)-based scheduling and staffing model	Produces cost-effective and efficient project schedules	Performance sensitive to problem domain characteristics	Hybrid optimization techniques and real-world case study validation
Rihm & Trautmann (2017)	Resource-constrained project scheduling	Novel MILP model with improved assignment and sequencing constraints	Outperforms existing models, especially under scarce resource conditions	Increased model complexity for large-scale instances	Further optimization for large datasets and integration with heuristic methods
Peralta et al. (2016)	Time and schedule management	Analysis of Personal Software Process (PSP) time records	Provides actionable recommendations for employee and manager schedule management	Context-specific findings tied to PSP adoption	Adaptation to agile and hybrid project management frameworks