Analysis for the implementation of a tool based on proportional hazard model

In these last years, both the focus towards which the maintenance activity has been directed, both companies general management context have undergone a remarkable evolution. On one hand, a growing level of demand and availability of the company’s assets, always taking into account the cost optimization philosophy; on the other, the development of ICT (Information and Communication Technologies), along with a Web service orientation, in an increasingly mobile and flexible scenario. In this setting, innovative solutions have emerged, in which new technologies have been applied, or existing solutions in various fields are combined in an unconventional way, leading to a compromise offering answers to all current requirements. As a specific example, predictive monitoring tools based on system conditions (predictive condition-monitoring tools and intelligent maintenance decision making), present a combination of statistical analysis and quality methods along with new monitoring and communication technologies. Precisely in a framework of such instruments, this thesis can be contextualized. The purpose of this thesis is to develop, through the implementation of theoretical methods obtained after an analysis of the literature, two functional modules of a maintenance tool called ProDEST. ProDEST is an instrument that is currently under development by the LGM (Laboratorio di Gestione della Manutenzione) of DIG (Department of Management, Economics and Industrial Engineering of Politecnico di Milano) and that basically consists in a tool that, through a series of independent modules offers a set of features capable of supporting maintenance activity. In order to provide a current solution, ProDEST is willing to be placed in the context of Web-based services and remotely managed applications. For that, a specific hardware solution was selected, (i.e National Instruments Hardware). Consequently, due to compatibility reasons with such hardware, programming with Labview language is imposed as a bond in this thesis. Within this project, the main goals assigned to this work have been: In the first place, to implement a software function based on PHM (Proportional Hazard Model). This method, which has already had a long history in different applicative fields, allows to obtain the risk level of the system, taking into account a series of key factors that accurately represent its state. The starting points have been previous works, in which the calculation of this important parameter for reliability studies, was obtained in a simplified case where only one of the parameters of the system was considered (mono-covariate case). Regarding this, the targets were: first of all, implementing once again the single parameter case (mono-covariate case), this time using a programming language compatible with the proposed tool, such as Labview. Secondly, the implementation, always provided in Labview language, of the general case in which various parameters can be used to describe the state of the system (multi-covariate case). The second objective is to offer an economic model that may be implemented with a programming language, in order to provide results that would once again help the maintenance expert in taking decisions. Specifically, this module is able to establish a threshold level of risk which, basing on economic factors, determines whether the maintenance operation is convenient or not. Furthermore, this model offers, always from an economic point of view, an optimal maintenance policy, which proposes various moments to perform the maintenance activity, depending on the degradation level of the system.