RPPS Research Paper | GitHub Static Page

PDF Document Showcase

Attached Project Documents

This section showcases the initially attached RPPS documents. Upload these PDF files inside your GitHub project under the documents/ folder and keep the file names same as used below.

Research Paper

RPPS research paper covering abstract, indexed terms, introduction, algorithms, methodology, findings, conclusion, and references.

Open Research Paper

Literature Review

Literature review covering workforce analytics, resource constraints, root cause analysis, risk analysis, supply chain model, and conceptual framework.

Open Literature Review

Final Project Report

Final report including summary, objectives, dataset description, methodology, analysis, findings, managerial implications, and future scope.

Open Final Report

Author & Project Information

Project Title: Resource Project Planning System (RPPS)

Author: Akansha Rana

Department: Department of Data Science, MSC

Institute: Vellore Institute of Technology

Email: akansha.rana2024e@vitstudent.ac.in

Approver: Dr. Murali S

Research Paper Section

Research Paper

Title

Resource Project Planning System: Enhancing Workforce Utilisation and Project Efficiency through Intelligent Planning

Abstract

In modern IT, government, consulting, and service-based organisations, workforce underutilisation is a major operational challenge. Employees who remain on the bench without active project allocation increase operational costs, reduce productivity, and create skill stagnation. The Resource Project Planning System (RPPS) is proposed as a data science-driven workforce optimisation framework that integrates machine learning, recommendation systems, optimisation algorithms, risk analysis, forecasting, and business intelligence dashboards.

RPPS analyses candidate profiles, project requirements, employee skills, experience level, domain expertise, performance records, and workforce demand. The system provides automated candidate-project matching, skill-gap analysis, global resource allocation, candidate segmentation, performance tracking, bench-risk monitoring, and future demand forecasting.

Indexed Terms

Workforce Analytics Resource Allocation Bench Management Skill Gap Analysis Project Recommendation System Machine Learning in HR Predictive Modelling Talent Optimisation

1. Introduction

Organisations frequently face resource planning issues such as wrong allocation, poor bench visibility, project staffing delays, skill gaps, and weak forecasting. Traditional systems often depend on manual spreadsheets, resulting in delayed decision-making and mismatched allocation. RPPS converts workforce planning from a reactive process into a proactive, data-driven decision system.

2. Problem Statement

The major problem addressed by RPPS is the inability of organisations to efficiently map available talent to suitable project opportunities. This leads to bench cost, underutilised employees, skill-demand mismatch, and project delivery delays.

3. Research Objectives

Develop an automated data pipeline for candidate and project data processing.
Build a recommendation engine for candidate-project matching.
Implement global resource allocation using optimisation algorithms.
Perform candidate segmentation using clustering techniques.
Track workforce performance using productivity KPIs.
Forecast future resource demand and workforce availability.
Provide actionable insights through interactive dashboards.

Literature Review Section

Literature Review

The literature review identifies major constraints in workforce and project resource planning. These constraints include human capital constraints, economic constraints, operational constraints, and knowledge or IT constraints. RPPS addresses these constraints using a combined analytical approach.

Comparative Analysis: Resource Constraints and Algorithmic Interventions

Constraint Category	Recommended Algorithm	Advantages	Limitations
Skill Matching Human Capital	Cosine Similarity, NLP, TF-IDF	Effective for semantic matching between candidate profiles and project descriptions.	Requires clean and standardised skill data.
Resource Assignment Operational	Hungarian Algorithm	Finds optimal one-to-one project-candidate assignment.	Can become computationally expensive for very large datasets.
Utilisation Prediction Bench / PIP Risk	Random Forest, XGBoost, Logistic Regression	Identifies risk profile of employees likely to remain on bench.	Some models may be difficult to explain to HR teams.
Demand Forecasting Financial Planning	ARIMA, Linear Regression, Prophet	Forecasts seasonal hiring and future resource demand.	Requires historical data for higher accuracy.
Workforce Segmentation Capability Planning	K-Means Clustering	Groups employees into meaningful workforce segments.	Selecting the correct number of clusters can be subjective.

Research Gap

Existing workforce planning systems usually focus only on staffing or scheduling. Very few systems integrate recommendation, optimisation, forecasting, performance tracking, skill-gap analysis, risk analysis, and dashboarding into one intelligent framework. RPPS fills this gap by combining workforce analytics and machine learning into a single planning system.

Root Cause Analysis

RPPS identifies the root cause of resource underutilisation as the absence of an integrated system that connects historical workforce data, predictive demand, and prescriptive skill matching.

People

Skill gaps, PIP isolation, low motivation, and mismatch between expertise and demand.

Process

Reactive staffing, fragmented planning, manual mapping, and poor forecasting.

Technology

No analytics dashboard, no skill mapping tools, and disconnected data sources.

Final Project Report Section

Project Report

Executive Summary

RPPS was developed as an intelligent workforce optimisation platform that integrates data engineering, machine learning, optimisation algorithms, forecasting models, and business intelligence tools to automate candidate-project matching and improve workforce utilisation.

System Components

Data Pipeline

Collects, cleans, validates, and prepares candidate and project data.

Recommendation Engine

Uses TF-IDF and cosine similarity for project matching.

Global Allocation

Uses Hungarian Algorithm for optimal resource assignment.

Segmentation

Uses K-Means to group candidates based on skills and capabilities.

Performance Tracking

Tracks execution accuracy, completion efficiency, and performance yield.

Forecasting

Uses Linear Regression and ARIMA to forecast workforce demand.

Dataset Description

Candidate Dataset

Candidate ID
Candidate Name
Email
LinkedIn
Skills
Technologies
Experience Level
Domain
Profile Description

Project Dataset

Project ID
Project Title
Project Role
Required Skills
Required Experience
Domain
Budget
Company Name
Project Description

Performance Dataset

Candidate ID
Project ID
Execution Accuracy
Completion Efficiency
Task Complexity
Performance Yield
Date

Methodology

Phase 1: Data Engineering Pipeline

Candidate profiles, project requirements, and performance records are collected and transformed into structured datasets. Data cleaning includes missing value treatment, duplicate removal, text standardisation, and feature validation.

Phase 2: Feature Engineering

candidate_soup = skills + tools_technologies + profile_description + experience_level

project_soup = required_skills + tools_required + project_description

data_soup = candidate_soup + project_soup

Phase 3: Recommendation Engine

TF-IDF converts text data into numerical vectors. Cosine similarity compares candidate profiles with project requirements.

TF-IDF = TF(t,d) × IDF(t)

Cosine Similarity = (A · B) / (||A|| × ||B||)

Phase 4: Global Allocation Engine

The Hungarian Algorithm is used to find the globally optimal assignment between candidates and projects.

Z = min Σ Σ Cij Xij

Phase 5: Candidate Segmentation

K-Means clustering groups candidates into meaningful clusters such as high-skill candidates, moderate-skill candidates, specialised experts, and candidates requiring upskilling.

Phase 6: Forecasting Engine

Linear Regression and ARIMA models forecast daily, weekly, monthly, and yearly resource demand.

Risk Analysis

Risk analysis in RPPS acts as a proactive workforce health check. It identifies bench stagnation, skill obsolescence, PIP failure risk, and project delay risk.

Risk Factor	Algorithm	Metric	Actionable Strategy
Bench Stagnation	Logistic Regression	Deployment Probability < 0.3	Automated Upskilling Trigger
Skill Obsolescence	TF-IDF Comparison	Skill Relevance Index	Strategic Training Shift
PIP Failure	Random Forest	Risk Score > 0.8	Structured Mentorship Assignment
Project Delay	Monte Carlo Simulation	Probability of Completion	Resource Reallocation from Bench

Supply Chain Model for Workforce Planning

RPPS treats workforce planning like a supply chain model. The bench is considered workforce inventory, upskilling is the transformation process, and project demand is the final requirement.

Supply

Internal talent pool and external hiring pipeline.

Inventory

Bench, PIP, and underutilised employees.

Transformation

Upskilling and readiness improvement.

Demand

Live and forecasted project requirements.

EOQ = √(2DS / H)

Where D is annual demand for a skill, S is training setup cost, and H is the holding cost of a resource on the bench.

Conceptual Framework: RPPS IPO Model

Input

Candidate data, project data, skills, experience, performance logs, bench records, PIP records, and risk indicators.

Process

TF-IDF, cosine similarity, Hungarian Algorithm, K-Means clustering, Random Forest, Logistic Regression, and ARIMA.

Output

Candidate recommendations, allocation matrix, skill-gap analysis, risk alerts, forecasts, and dashboards.

Key Findings

Skill-based recommendation improves candidate-project matching.
Global optimisation reduces allocation conflicts.
Candidate segmentation supports targeted workforce planning.
Performance tracking improves productivity visibility.
Forecasting supports future resource planning.
Risk analysis helps identify bench stagnation and PIP failure early.
Dashboards support data-driven managerial decision-making.

Managerial Implications

RPPS helps HR managers, project managers, and leadership teams reduce bench cost, improve internal deployment rate, identify skill gaps, design upskilling programs, forecast future demand, and make strategic workforce decisions. It is applicable in IT services, consulting firms, staffing agencies, healthcare organisations, manufacturing, education, and government organisations.

RPPS Static Profile Builder

This profile builder allows users to create a static candidate profile and generate a candidate soup for RPPS demonstration.

Candidate Name

Email

Primary Category

Sub Role

Experience Level

Skills

Tools & Technologies

Profile Description

References

Resource Project Planning System Literature Review, Department of Data Science, VIT.
Resource Project Planning System Final Project Report, Department of Data Science, VIT.
Research methods related to workforce analytics, recommendation systems, optimisation algorithms, clustering, and forecasting.
TF-IDF, Cosine Similarity, Hungarian Algorithm, K-Means Clustering, Random Forest, Logistic Regression, ARIMA, and Business Intelligence dashboarding concepts.