Project Analytics : A Learning Based Approach

E-Book Overview

Abstract : Classical models of resource constrained project scheduling problems are not adequate to solve real world problems due to increased complexities and uncertainties. Intelligent project analytics are essential for complex, fuzzy, stochastic, multi-mode, resource constrained project scheduling problems with multiple objectives. This work explores how to apply the concept of intelligent deep analytics for project management. Efficient project management requires coordination and integration among seven elements associated with a project such as Scope (novelty, objectives, constraints), System (technology), Structure (complexity), Staff, Skill (innovation, design, SCM, ERP) &, Style (pace, leadership), Security (threat analysis, risk assessment and mitigation) and Strategy (shared vision, communication). This work presents an algorithmic Project Analytics Mechanism (PAM) in terms of agents, input, output, strategic moves, case based planning algorithm, performance metrics, revelation principle, verification protocols for security intelligence and payment function. The intelligence of PAM is explored through a set of strategic moves such as case based planning, collaborative, security and collective intelligence. The complexity of the analytics is analyzed in terms of computational cost and security analysis. Traditionally, the computational burden of project planning depends on the efficiency of heuristic search algorithm to find out the critical path of a project. But, it may not capture the uncertainties, risks and complexities involved in a real world project. The computational complexity of PAM is associated with the efficiency of case based reasoning (CBR) i.e. case retrieval and case adaptation algorithms. Case based planning searches reference plan from a case base through efficient case retrieval and adaptation mechanism. 100% matching in case retrieval is a NP hard problem. Traditionally, many CBP algorithms have tried to find exact matching between the graphs of resource and time constrained project network. It may be practically infeasible. The basic objective of K-Nearest Neighbor Search algorithm in PAM is to search for approximate matching among the neighbors. The project analytics monitor project performance and adjusts the reference plan. The revelation principle preserves the privacy of contracts and payment function through signcryption. This work also outlines the architecture of an intelligent project analytics in terms of computing, communication, data, application and security schema. The concept of deep project analytics and PAM has been applied to analyze three test cases – smart village project, smart city project and software project management. Keywords : Project analytics, Deep Analytics, PAM, Case based planning, Case retrieval, Case adaptation, Smart villages, Smart cities, Software project management

E-Book Content

Project Analytics : A Learning based Approach Sumit Chakraborty Fellow, MIS (Indian Institute of Management Calcutta), Bachelor of Electrical Engineering (Jadavpur University), India. E-mail: [email protected], [email protected]; Phone: 91-9940433441 Abstract : Classical models of resource constrained project scheduling problems are not adequate to solve real world problems due to increased complexities and uncertainties. Intelligent project analytics are essential for complex, fuzzy, stochastic, multi-mode, resource constrained project scheduling problems with multiple objectives. This work explores how to apply the concept of intelligent deep analytics for project management. Efficient project management re
You might also like

Money Management Strategies For Futures Traders
Authors: Nauzer J. Balsara    216    0


The Compleat Day Trader
Authors: Jake Bernstein    254    0


Fibonacci Forecast Examples
Authors: Emmett T.J.    149    0


Economics Of Mobile Telecommunications
Authors: Harald Gruber    200    0


Developer To Designer: Gui Design For The Busy Developer
Authors: Mike Gunderloy    187    0


Project Management
Authors: Gary R. Heerkens    165    0


Statistical Process Control
Authors: John S Oakland    151    0