Continuous integration (CI) is a widely used practice in modern software engineering. Unfortunately, it is also an expensive practice — Google and Mozilla estimate their CI systems in millions of dollars. In this dissertation, I propose a collection of novel build selection approaches that are able to save the cost of CI. I also propose the first exhaustive comparison of techniques to improve CI including build and test granularity approaches. I firstly design a build selection approach (SMARTBUILDSKIP) for CI cost reduction in a balanceable way. The evaluation of SMARTBUILDSKIP shows that it can save a median of 30% of builds by only incurring a median delay of 1 build in a median of 15% of failing builds under its most conservative configuration. To minimize the delayed failure observation, I then propose the second build selection approach (PRECISEBUILDSKIP) that can save cost without delaying failure observation. We find that PRECISEBUILDSKIP can save a median of 5.5% of builds while capturing the majority of failing builds (100% in median) from the evaluation. After that, I evaluate the strengths and weaknesses of 10 techniques that can improve CI including SMARTBUILDSKIP. The findings of the comparison motivate my next work to design a hybrid technique (HYBRIDBUILDSKIP) that combines these techniques to produce more cost-saving while keeping a low proportion of failing builds that are delayed in observation. Finally, I design an experiment to understand how different weights of test duration among the whole build duration can influence the cost-saving of build and test selection techniques. / Doctor of Philosophy / Modern software developing teams commonly use the continuous integration as the practice of automating and testing the integration of code changes from multiple contributors into a single software project. The best practice of continuous integration requires this process happens as frequently as possible because the bugs can be found earlier and easier before the change sets grow too large. However, continuous integration process can be time-consuming and in most cases the code change is bug-free. This means that developers may have to wait for a long time only to get a result with no actionable feedback. Thus, in this dissertation, I present multiple selection approaches to selectively execute the continuous integration process based on the prediction of the outcome - if the outcome is predicted to be passing with no actionable feedback, the approach will decide to skip the current execution. The evaluation result shows that my approaches can save the cost of continuous integration while keeping the value of it (finding bugs earlier).

The maintenance cost for an engineering system is an uncertain quantity due to uncertainties associated with occurrence of failure and the time taken to restore the system. The problem of probabilistic analysis of maintenance cost can be modeled as a stochastic renewal-reward process, which is a complex problem. Assuming that the time horizon of the maintenance policy approaches infinity, simple asymptotic formulas have been derived for the failure rate and the cost per unit time. These asymptotic formulas are widely utilized in the reliability literature for the optimization of a maintenance policy. However, in the finite life of highly reliable systems, such as safety systems used in a nuclear plant, the applicability of asymptotic approximations is questionable. Thus, the development of methods for accurate evaluation of expected maintenance cost, failure rate, and availability of engineering systems is the subject matter of this thesis. In this thesis, an accurate derivation of any m-th order statistical moment of maintenance cost is presented. The proposed formulation can be used to derive results for a specific maintenance policy. The cost of condition-based maintenance (CBM) of a system is analyzed in detail, in which the system degradation is modeled as a stochastic gamma process. The CBM model is generalized by considering the random repair time and delay in degradation initiation. Since the expected cost is not informative enough to estimate the financial risk measures, such as Value-at-Risk, the probability distribution of the maintenance cost is derived. This derivation is based on an interesting idea that the characteristic function of the cost can be computed from a renewal-type integral equation, and its Fourier transform leads to the probability distribution. A sequential inspection and replacement strategy is presented for the asset management of a large population of components. The finite-time analyses presented in this thesis can be combined to compute the reliability and availability at the system level. Practical case studies involving the maintenance of the heat transport piping system in a nuclear plant and a breakwater are presented. A general conclusion is that finite time cost analysis should be used for a realistic evaluation and optimization of maintenance policies for critical infrastructure systems.

The maintenance cost for an engineering system is an uncertain quantity due to uncertainties associated with occurrence of failure and the time taken to restore the system. The problem of probabilistic analysis of maintenance cost can be modeled as a stochastic renewal-reward process, which is a complex problem. Assuming that the time horizon of the maintenance policy approaches infinity, simple asymptotic formulas have been derived for the failure rate and the cost per unit time. These asymptotic formulas are widely utilized in the reliability literature for the optimization of a maintenance policy. However, in the finite life of highly reliable systems, such as safety systems used in a nuclear plant, the applicability of asymptotic approximations is questionable. Thus, the development of methods for accurate evaluation of expected maintenance cost, failure rate, and availability of engineering systems is the subject matter of this thesis. In this thesis, an accurate derivation of any m-th order statistical moment of maintenance cost is presented. The proposed formulation can be used to derive results for a specific maintenance policy. The cost of condition-based maintenance (CBM) of a system is analyzed in detail, in which the system degradation is modeled as a stochastic gamma process. The CBM model is generalized by considering the random repair time and delay in degradation initiation. Since the expected cost is not informative enough to estimate the financial risk measures, such as Value-at-Risk, the probability distribution of the maintenance cost is derived. This derivation is based on an interesting idea that the characteristic function of the cost can be computed from a renewal-type integral equation, and its Fourier transform leads to the probability distribution. A sequential inspection and replacement strategy is presented for the asset management of a large population of components. The finite-time analyses presented in this thesis can be combined to compute the reliability and availability at the system level. Practical case studies involving the maintenance of the heat transport piping system in a nuclear plant and a breakwater are presented. A general conclusion is that finite time cost analysis should be used for a realistic evaluation and optimization of maintenance policies for critical infrastructure systems.

Thesis (Ph. D.)--University of Wisconsin--Madison, 1974. / Typescript. Vita. eContent provider-neutral record in process. Description based on print version record. Includes bibliography.

Railway transportation plays a vital role in modern societies. Due to increasing demands for transportation of passengers and goods, higher speed trains with heavier axle loads are introduced to the railway system, and it is expected to continue in the future. Therefore, track geometry bears huge static and dynamic forces that accelerate degradation process. As a result, railway track should be inspected regularly to detect geometry faults and to plan for maintenance actions in advance. Track geometry inspection has a profound impact on railway track availability and maintenance cost. Although there have been improvements in safety performance and maintenance planning of railway tracks, still infrastructure managers expect a more effective maintenance planning and scheduling regime. This thesis proposes a simulation-based model for optimization of track geometry inspection intervals. To simulate the track geometry evolution a linear model is used to model track geometry degradation in a maintenance cycle. It is assumed that the parameters of degradation model are random variables following lognormal distribution. Using the proposed model, the track geometry behaviour is simulated under different inspection intervals. Later, different inspection intervals are compared with respect to the cost function and the optimal range of inspection intervals is obtained.

Le contexte hospitalier actuel exhorte les établissements de santé à une gestion plus rigoureuse et objective des ressources dans le but de maitriser leur performance. Le patrimoine immobilier des établissements publics de santé contribue à la qualité des soins demandée, à la qualité d'accueil des patients et des conditions de travail du personnel. Aujourd'hui, les hôpitaux demeurent conçus et réalisés sans aucune anticipation des coûts futurs d'exploitation-maintenance. Ces coûts qui représentent indéniablement une valeur importante du budget global de fonctionnement, sont pourtant méconnus. Afin de palier à la faiblesse des moyens de financement public, l'administration a comme recours la mise en place d'une nouvelle procédure de contrats de Partenariat Public Privé. Or celle-ci ne dispose pas d'un minimum d'informations en terme de connaissances sur les dépenses d'exploitation-maintenance actuelles de son patrimoine. Dans ce contexte, le coût global constitue une approche innovante dans la gestion des projets immobiliers.La problématique développée dans ce travail de thèse est donc d'approfondir les connaissances en amont des coûts d'exploitation-maintenance des bâtiments hospitaliers. Elle apporte un éclairage sur la relation entre les coûts d'exploitation-maintenance et l'activité hospitalière. Elle propose ainsi une réponse plus adaptée et plus opérationnelle à la question de l'estimation de ces coûts d'exploitation-maintenance en aidant le maître d'ouvrage à prendre en compte l'activité de soin et les choix architecturaux dans son projet. Cela revient à proposer une méthode d'aide à la décision pour construire un bâtiment hospitalier qui minimise les dépenses futures de fonctionnement. Cette méthode s'appuie sur un modèle d'estimation et sur un observatoire des coûts d'exploitation-maintenance réel de 37 établissements publics de santé. / The current situations of hospitals require more efficient management in order to control their performance. The real estates of public hospitals contribute to the quality of required healthcare, the quality of patient care and working conditions of healthcare staff.Today, hospitals are designed and built without anticipating of future evolution of operation and maintenance costs. These costs that are an important contribution to the value of the overall operating budget are underestimated and far unknown.To overcome the lack of public finance, the government has so used the Public Private Partnership Contracts as a new procedure. Yet the government does not have a minimum of information about operating and maintenance expenses of its real estates. In this context, the Whole Life Cycle Costing is an innovative approach in managing public healthcare real estate projects.The problem developed in this thesis is to expend is to expend the knowledge of operation and maintenance costs of healthcare real estates. This thesis highlights the relationship between operating and maintenance costs and hospital activity. It proposes a more efficient response in the issue of estimating these costs by helping the director to take into account the healthcare activity and the architectural choices. It provides a decision making tool for designing and building a new hospitals for minimizing future costs of operation.This method is based on a model for estimating costs and on an observatory of real operation and maintenance costs of 37 public hospitals.

Winter in Northern Sweden comes with very harsh and unpredictable condition associated with large amounts of snowfall covering roadways thereby affecting transportation by roads. When the road conditions i.e. the snow depth, road unevenness and friction of the road surface are accessed and found to exceed the threshold, a maintenance action must be carried out to retain the road to the required condition for the user. The aim of maintenance, in this case, is to make the road comfortable, safe and economical for the road user. Decision support system, therefore, comes in handy to facilitate on deciding what maintenance action to carry out and when the action should be carried out, where the action should be carried out and how to go about the action based on the various data and resources available. This thesis project concentrates on how to carry out a winter road maintenance after receiving an alert of an action to carry out, when to carry it out and the road network that needs to be maintained. The thesis work focusses only on two of the winter road maintenance actions namely snow ploughing of bus stops in Luleå and application of abrasives commonly referred to as sanding of bus stops. Carrying out winter road maintenance comes at a huge cost from both direct and indirect costs with the Swedish government spending about SEK 1.75 billion every year as indicated by Jana Sochor and Cecilia Yu (Sochor & Yu, 2004). This means that reduction in the maintenance cost of even 5% through optimisation of the maintenance cost would translate into a saving of about 87.5 million SEK per year and in 10 years could amount to close to 1 billion SEK. Optimization also leads to efficiency and effectiveness that could result in improved movement on the road and reduced environmental and social-economic impacts. Maintenance planning thus becomes essential for the effective and efficient execution of work and utilisation of the available scarce resource. This thesis project focusses on the use of Operations research methods to minimise the cost of carrying out a winter road maintenance action by finding the near optimal or if possible optimal solution and still deliver the required service level. The thesis delivers two main things: It first delivers a framework to support winter road maintenance decision making after an alert of an action is received and secondly an algorithm for the route that minimises the cost of maintenance by providing the route that minimises the travel distance of the ploughing/sanding vehicle from its source depot and back to the depot after completing a maintenance action assuming that the vehicle and material (fuel and sand) are in the same depot. The routes with minimum travel distance will, therefore, be that route that will reduce the labour time and in turn the labour cost, reduce the fuel consumption and the maintenance of the equipment due to reduced usage. The project uses a vehicle routing problem which is a generalised travelling Salesman as the optimisation technique to determine the optimal solution for the allocation of resources for carrying out a maintenance action to facilitate efficient utilisation of the available resources. This is with the help of a commercial optimisation software and support tools namely ArcGIS. To come up with the algorithm, the first step was a digital representation of the vehicle road network in Luleå for network analysis after which the bus stops were imported from google earth into the network. A two-stage optimisation was then carried out: first was a model for route optimisation based on the road network in ArcGIS with the objective function to minimise the travel distance and constraints based on the available resources. The results of ii the model were then exported into excel for the second optimisation for the optimal cost of maintenance done through a developed excel algorithm. The total cost of maintenance comprised direct and indirect cost. The direct cost consisted of the cost of fuel, the cost of personnel and the cost of hiring vehicles while the indirect cost results from the penalty fee charged for sanding and ploughing a bus stop after the threshold time given to a maintenance contractor by the municipality. Any bus stop that is ploughed after the threshold attracts a penalty per hour of the exceeded time. Six penalty threshold times were considered i.e. 30, 60, 90, 120, 150 and 180 minutes and a single parameter deterministic sensitivity analysis was carried out for each cost parameters to determine the sensitivity of the total maintenance cost. The more relaxed penalty thresholds were found to be less sensitive to the direct cost and the total maintenance cost compared to the more sensitive ones. When the penalty threshold is relaxed, the optimal maintenance cost reduces, and the required number of vehicles reduces. The cost of vehicle hire was found to be more sensitive than the other costs. The results of this project can help the maintenance contractor in developing a work schedule for the maintenance personnel and improve vehicle fleet management. By modelling the worst scenario, a contractor can plan for the maximum number of vehicles required and consequently the personnel required. With the optimal travel route for each vehicle and the total maintenance cost determined, maintenance contractors can determine the sustainability and profitability of their business and be able to negotiate for a better and more sustainable agreement (Contract) or for the relaxation of the penalty threshold time if it does not affect the service level required i.e. the quality and safety requirements. The approach used in this project can also be used for other winter road maintenance problems.

Utbyggnaden av spårvägen Linje 7 i Stockholm som ska utmynna i Spårväg City har tagit sin början. Vilka typer av spårvägskonstruktioner som ska användas på alla de ingående sträckorna är däremot inte bestämt. Grässpår är ett alternativ, men kunskapen om grässpår är inte stor i Stockholmsregionen på grund av att denna spårvägstyp tidigare inte projekterats i regionen. Denna rapport ska underlätta valet genom informationshämtning från intervjuer, möten, mejlkontakt, studiebesök och litteraturstudier. Grässpåret jämförs med gatuspår i asfalt som används i Spårväg City. Ballastspår på egen banvall är den enklaste spårvägskonstruktionen, och används i jämförelserna. Dock kommer den konstruktionen inte att användas på Spårväg City. De olika spårvägskonstruktionerna jämförs utifrån anläggningskostnader samt underhållskostnader för de olika konstruktionstyperna. En grundlig jämförelse har även gjorts byggtekniskt. Vid jämförelsen av de olika konstruktionerna så var gatuspår i asfalt den dyraste att anlägga, cirka 20 000kr/m. Det var mer än 3 gånger så dyrt som grässpåret med en kostnad på cirka 6000kr/m, som i sin tur var marginellt dyrare än ballastspåret på egen banvall med en kostnad på cirka 3000kr/m. Även underhållskostnaden var betydligt högre för gatuspåret i asfalt. Gatuspåret i asfalt var nästan dubbelt så dyrt som grässpåret att underhålla. Grässpåret i sin tur var lite mer än dubbelt så dyrt som ballastspåret på egen banvall att underhålla. Gatuspåret i asfalt fick en underhållskostnad på cirka 160 000 kr/km & år, grässpåret fick en underhållskostnad på cirka 85 000kr/km & år, och för ballastspår på egen banvall cirka 40 000kr/km & år. / The extension of the tramway line 7 in Stockholm, which will culminate in “Spårväg City”, has begun. Which types of tramway constructions to use on all the included distances has not been decided. Grass Track is an option, but there’s a lack of knowledge of the grass track in Stockholm because this type of tramway has not been projected before in the region. This report will simplify the selection, by information extraction from interviews, meetings, e-mail contact, field trips and literature studies.  The grass track will be compared with street track in asphalt which is used in “Spårväg City”. Ballast track on its own embankment is the simplest tramway construction, and is used in the comparisons. However, the design is not to be used on “Spårväg City”. The various tramway constructions are compared by the construction costs and the maintenance costs for the different construction types. A thorough comparison has also been made in the construction technical part.  When comparing the different constructions, it was the street track in asphalt that was the most expensive to construct about 20 000kr/m. It was more than 3 times as expensive as the grass track with a cost of about 6000kr/m, which in turn was marginally more expensive than the ballast track on its own embankment with a cost of about 3000kr/m. Even the maintenance cost was significantly higher for street track in asphalt. The street track in asphalt was almost twice as expensive as the grass track to maintain. The grass track in turn was a little more than twice as expensive as the ballast track on its own embankment to maintain. The street track in asphalt had a cost of about 160 000kr/km & year, the grass track had a cost of about 85 000kr/km & year, and the ballast track on its own embankment had a cost of about 40 000kr/km & year.

Thesis (MBA)--University of Stellenbosch, 2010. / The City of Cape Town water services is currently operating on the reactive mode and their maintenance system is not effectively and efficiently utilized. The lack of proper utilization of Physical Asset Management (PAM) has resulted in the technical department failure to meet the departmental objectives of providing planned and scheduled maintenance activities, reducing equipment downtime and ensuring planned delivery of quality spares at the appropriate time. The aim of the report is to provide guidance in the implementation of physical assets management. The report looks at different maintenance strategies available and evaluates the current operating scenarios for the department and further recommends the relevant strategy that will suit the department. Physical asset management as a maintenance program provides an effective planned and scheduled maintenance system to reduce labor and downtime. This ensures that the correct materials and parts are utilized and that the workmanship is of a high quality. A Maintenance Plan provides documented and sequenced tasks with labour and types of material required to execute the plan. Infrastructure assets are systems that serve defined operational needs, where the intention is to maintain the asset for continual use on a certain level. One of the main aspects in infrastructure network is the degree of interdependency not only within a particular asset network but also among networks. The failure of one component within a network may undermine the performance of other networks. One of the major objectives in an infrastructure network is to maintain the necessary service level through continuous maintenance while ensuring cost effectiveness The initial step is to choose certain maintenance tactics and then decide how often these tactics will be performed. The frequency of maintenance, its actions or tactics depends on the current condition of the equipment which is linked within the technical characteristics of failure and specific monitoring technique. “Regardless of the system you select, you must start from where you are and then develop an orderly series of steps to get where you want to end up, so start by assessing your current situation. The most essential ingredient in your maintenance management system is the people. An excellent system run by poorly trained or unmotivated people will be adequate at best but well trained people with positive attitudes can make an excellent system world class.”-Thomas Westerkamp

The diploma thesis deals with the issues concerning implementation of the RCM (Reliability Centred Maintenance) method on large power installations (boilers and their accessories) with the use of the new AMST software tool in the conditions of a particular energy company.