Project Management through States

Abstract

Looking at computer science, rising complexities, durations and costs for software and hardware development make it necessary to employ a new holistic model to manage those development processes. This paper introduces a new model which is based on abstract states in order to calculate a project's complexity. Based on this complexity it is now possible to estimate key variables such as costs, duration and progress. An abstract definition of states allows users to adjust this model to their project and to their specific requirements. I. INTRODUCTION A. MotivationProject management has become a major issue for modernly organized companies. Most engineering disciplines exert well defined methods and models to manage, control and evaluate projects. But in computer science, the management of hardware and software projects is still very weak. This expresses not only in cost and time overruns but also in a high cancellation rate of such projects. For hardware development it is not enough to simply count transistors on integrated circuits anymore [1].The most common cost model in software development COCOMO (COnstructive COst MOdel) is based on lines of code [2]. Costs and duration of a project are adjusted by a questionnaire which determines its complexity. An empirical study of Chris Kemerer analyzed the accuracy of cost models on 15 large completed business data processing projects and has shown a mean error of 600% for projects planned with COCOMO [3].A better model enables project managers to set a tighter framework to a project. A good project management has to cover three basic purpose fields: a project should a) fulfill its requirements (appropriate), b) in the given time (in time) and c) with the given budget (cost effective) [4]. To achieve those goals, mathematical methods will be presented in the next sections which allow the determination of a project's relevant variables by using states. Furthermore, this model reduces the demand for empirical data. Today, most models base their estimations on empirical data [5]. Gathering this data is expensive and challenging. With fast changes in technology as well as with new abstraction layers, a data comparison from different projects is difficult [6].

Research and Measurement of Software Complexity Based on Wuli, Shili, Renli (WSR) and Information Entropy

“…On the basis of existing achievements [40], in the following two sections, information entropy is used to measure the complexity of the software project personnel organization hierarchy structure of and the complexity of personnel communication information. The complexity of systems can be measured by a methodology derived from Shannon's theorem [41][42][43][44][45][46].…”

Quantitative Analysis of Software Code by States

“…The set of states is the abstraction of a real system" [15]. This method was used to calculate the complexity of in hardware designs [14] [13]. Thereby, the complexity was not estimated anymore, but calculated [14].…”