Att upprätthålla funktionaliteten på ett fungerande tekniskt system, som användes av människor eller organisationer för ett specifikt ändamål, är alltid viktigt. Om ett system plötsligt och oväntat slutar fungera kan det ha stora negativa konsekvenser. Att vänta tills ett system går sönder och reparera sedan är inte en framgångsrik strategi, systemet måste underhållas så att fel inte inträffar. Men fortlöpande regelbundet underhåll kan vara resurskrävande, det som behövs är att systemet kontinuerligt övervakas och underhålls när övervakningssystemet bedömer att det behövs. Denna studie syftar till att studera konceptet ’preventivt underhåll’ (Preventive Maintenance) för att se vilka utmaningar och möjligheter det finns vid användning av uppkopplade enheter (Internet of Things, IoT). Arbetet genomfördes genom intervjuer av företag och forskare, studium av befintliga lösningar från industrin, och jämförelse med tidigare forskning. Resultatet visar att det råder en viss okunskap hos företag och akademi för vad IoT-baserade lösningar kan åstadkomma. / To maintain the function on a technical system, that is being used by people or organisations for a specific purpose, is of important value. It can have huge negative impact on business, if a system suddenly and unexpectedly stops working. To wait until a system goes down and to repair it immediately is not a successful strategy. The system need maintenance. Though continuous regularly maintenance can be resource demanding. What is needed are continuously monitored systems. This study aims to study the concept of ’Preventive Maintenance’ to see which challenges and possibilities there are by using connected devices (Internet of Things, IoT). This work have been done through interviews from organisations and researches, study of existing solutions from the industry, and to object those against previous research. The results of this study shows that there is a slight lack of knowledge from organisations and academy for what these solutions can accomplish.

A quasi-renewal process is more realistic in modeling the behavior of a repairable system than traditional models such as perfect repair and minimal repair since it reflects the deterioration process of the system over time while traditional models do not. The quasi-renewal parameter is set to a value between 0 and 1 to indicate the rate of deterioration. Moreover, a quasi-renewal process can also be used to model the increasing time of maintenance actions due to the increasing difficulty of maintaining an aging system by setting the parameter to a value larger than 1. We construct a model where the operating times follow a quasi-renewal process and the corrective/preventive maintenance times follow another quasi-renewal process. A quasi-renewal function and two equivalent point availability expressions are developed for the model described by a quasi-renewal process with and age-dependent preventive maintenance policy. In addition, numerical results from various theoretical distributions are obtained to illustrate the behavior of the models. The two equivalent point availability functions each contains an infinite sum and must be truncated to obtain a numerical approximation. The two approximated point availability functions form upper and lower bounds on the real value. The bounds are useful for determining the result accuracy, which can be arbitrarily increased by adding more terms to the truncated summation. Our framework provides a new time-dependent availability model for a non-stationary process with a preventive maintenance policy without any cost structure or optimization problem. / Ph. D.

Organizações eficientes são aquelas que conseguem manter equilibradas as vertentes de qualidade, custo e tempo. Em relação ao último, existem várias etapas da cadeia produtiva nas quais o tempo deve ser monitorado. Quando a programação da produção nas indústrias não é priorizado, pode-se incorrer vários efeitos negativos. Um deles, é o atraso em relação à data de entrega, no qual a corporação pode sofrer penalidades financeiras, além de uma exposição negativa para a marca, a qual pode ter sua credibilidade contestada. Dessa forma, essa pesquisa tem por objetivo propor métodos construtivos, que minimize a medida de desempenho total tardiness (atraso total). Para aproximar o método à realidade vivenciada pelas indústrias, será considerada a restrição de manutenção preventiva. Além disso, o ambiente de estudo será o contexto de no-wait flowshop, no qual as tarefas são processados continuamente e sem que haja interrupções entre uma operação e outra de uma mesma tarefa. Além da proposição de métodos construtivos para a resolução do problema, apresenta-se uma metaheurística como forma de demostrar como pode-se aprimorar os resultados gerado pelos métodos construtivos. Experimentações computacionais foram elaboradas e realizadas para comparação dos algoritmos. Dentre as heurísticas construtivas a que apresentou melhor desempenho foi a \"EDD + NEH + LS1 + LS2\'\', na qual utiliza uma lógica de inserção. A metaheurística proposta é baseada no procedimento IG (iterated greedy), sendo que há melhora de resultado em relação as heurísticas construtivas. Assim, espera-se que essa pesquisa possa ser utilizada e aplicada pela indústria de manufatura para aumentar a efetividade da programação da produção. / Efficient organizations are those that manage to keep the quality, cost and time strands balanced. With respect to the variable time, there are several stages of the production chain in which it must be monitored. When scheduling in companies is not prioritized, several negative effects incur. One of them is the delay in relation to the due date, for which the corporation can suffer financial penalties, in addition to a negative exposure to the brand, which may have its credibility challenged. Therefore, this research aims to propose constructive methods, which minimizes the performance criterion of total tardiness. In order to approximate the method to the reality of the industries, preventive maintenance constraints will be considered. And the environment of the study will be the no-wait flowshop, in which jobs are processed continuously and without interruptions between one operation and another of the same job. In addition to proposing constructive methods to solve the problem, a metaheuristic is presented as a way of demonstrating how to improve the results generated by the constructive methods. Computational experiments were elaborated and performed for comparison of the algorithms. Among the constructive heuristics that presented the best performance was \"EDD + NEH + LS1 + LS2\", in which it uses an insertion logic. The proposed metaheuristic is based on the IG (iterated greedy) procedure, and there is an improvement of the result in relation to the constructive heuristics. Thus, it is expected that this research can be used and applied by the manufacturing industry to increase the effectiveness of scheduling.

La qualité et la fiabilité des infrastructures routières jouent un rôle majeur dans la sécurité routière. Cela est d'autant plus vrai lorsque qu'on s'intéresse à la circulation de véhicules autonomes qui doivent être capables à terme de circuler seuls dans l'environnement routier. Les récents travaux menés au sein de VEDECOM montrent qu'une signalisation horizontale routière claire et visible est importante dans sa prise de décision. La détection des lignes de marquage sont en grande partie réalisée par des caméras. Afin d'optimiser cette approche et prévenir les situations de non détection, cette thèse propose des outils d'analyse de fiabilité et d'aide à la maintenance de la signalisation horizontale. Aujourd’hui, la fiabilité des marquages est basée le phénomène de rétro-réflexion : un véhicule éclaire avec ses feux de croisement un marquage qui renvoie la lumière vers le conducteur. Pour établir son niveau de dégradation, des campagnes d'inspection sont réalisées par des rétro-réflectomètres montés sur des véhicules traceurs. La littérature des trente dernières années présente essentiellement des modèles de dégradation basés sur des méthodes de régression. Ces derniers présentent de nombreuses difficultés à être déployés dans le cadre d'un plan de maintenance. Cette thèse propose d'aborder ces questions sous l'angle de la théorie de la fiabilité et de la maintenance tout en tenant compte des pratiques actuelles. Une ligne de marquage est ici interprétée comme un système multi-composants monté en parallèle. Cette thèse propose de l'analyser en quatre points. Premièrement, l'ensemble des inspections est formalisé en une base de suivi. Si des données sont manquantes et si l'historique de maintenance est indisponible, alors différentes approches basées sur une Classification Ascendante Hiérarchique (CAH) sont proposées afin de les estimer. Deuxièmement, l'entretien de la totalité d'une ligne est logistiquement délicat. Une CAH de la base de suivi a pour fonction d'établir les marquages suivant un même modèle de dégradation. Les clusters sont géographiquement localisés et corrélés à des situations précises comme un échangeur ou une agglomération. Pour ces raisons, ils sont interprétés comme des zones de maintenance stratégiques. Troisièmement, réaliser une analyse de Weibull des marquages. Les rétro-réflectomètres n'indiquent pas précisément les instants de défaillance. Ils sont statistiquement censurés à gauche, à droite ou par intervalle. En alternative à la méthode du Maximum de Vraisemblance, une approche basée sur un algorithme EM est proposé afin d'établir le modèle de Weibull et les censures estimées les plus vraisemblables. Dernièrement, deux stratégies de maintenance sont proposées : systématique par rapport à l'âge et conditionnée par la dégradation courante. Elles sont en adéquation avec les pratiques de maintenance. La première permet une gestion passive de l'entretien tandis que la seconde permet une connaissance avancée de la ligne de marquage dans le temps. A partir d'un système multi-composants non réparable et fortement censuré, les composants suivant un même modèle de dégradation sont classables, chaque groupe connait un modèle de durée de vie et finalement il est possible de déduire un plan de maintenance adapté / The quality and reliability of road infrastructure and its equipment play a major role in road safety. This is especially true if we are interested in autonomous car traffic. Recent papers from VEDECOM Institut proves that a clear and reliable road marking is important in it decison making. Marking lanes are detected by camera. These markings need an accurate maintenance strategy to guarantee that the markings remain perceptible. This report proposes different solutions based on the reliabilty and maintenance theory. Today, the markings reliability is based on the retroreflective illuminance. A retroreflective marking reflects light from a vehicle headlight back in the direction of the driver. Marking retroreflectivity can be dynamically inspected using a retroreflectometer. The litterature of the last thirty years proposes degradation models for retroreflective marking based on a regression model. All of them have a common weakness: they are difficult to apply directly to a given road network. This report presents maintenance models who math with current maintenance actions. A marking lane is interpreted as multi-unit systeme. All unit are laid in parallel. The global maintenance strategy is based on four points. First, the whole inspection data is formalized into one monitoring base. If inspection data is missing or if the maintenance historic is unavailable else an estimation process based on the Agglomerative Hierarchical Clustering (AHC) is proposed. Second, to replace a whole markings lane is logistically difficult to work. Again, an AHC of the monitoring proposed several clusters. Each cluster presents it own degradation model. Clusters are geographically tracked and correlated to specific situation (interchange, urban area, bypass...). That's why a cluster is interpreted as a maintenance strategic area. Thirdly, a Weibull analysis of each cluster is done. Current retroreflectometers cannot detects the exact faillure moment. this information is statistically censored. Three cases are identified : left, right and interval censored. To parameter a Weibull model, an EM Algorithm is propoased as an alternative to the Maximum Likelihood Estimator. This algorithm is also an estimator to censored markings life time. Lastly, two classic preventive maintenance strategies are proposed : systematic according to the age and conditionned to the current degradation. Each one is credible according the current maintenance practice. The first prposed a passsive managament of the markings maintenance. The second ensures an advanced knowledge of the road network over the time. On a multi-unit system no-repairable and strongly censored, units which admit the same degradation model are identified by a clustering approach. Each cluster present it own Weibull analysis. Finally, an adapted maintenance strategy is done

Por estar diretamente sujeita a fatores econômicos de nível global, a mineração tende a se desenvolver para buscar processos produtivos mais enxutos, aliando menores custos operacionais a melhores taxas produtivas. Neste contexto, o objetivo principal deste trabalho é avaliar o comportamento geral da curva de tempo em manutenção preventiva de uma frota de caminhões em uma mineração de grande porte, sendo a análise realizada em uma janela temporal baseada em um período econômico de mínimo local nos preços do cobre. A partir do comportamento observado é feita uma discussão sob as premissas da metodologia de gestão nomeada Manutenção Produtiva Total de forma a se debater os possíveis principais riscos e/ou consequências que podem ser enfrentadas pelo projeto e como estas decisões qualitativamente indicam o grau de maturidade na integração operacional entre produção e manutenção em um projeto minerário. Ao final, pode-se concluir que ainda é necessário um intenso trabalho em se traduzir as premissas da MPT para a indústria mineral, buscando-se adaptar processos e métricas que sejam mais condizentes ao ambiente da indústria extrativa (que é muito diferente da indústria de manufatura, onde o programa teve início). Além disso, fica claro que a aplicação da MPT é possível e bem-vinda na mineração já havendo de vários estudos acadêmicos na área - ou seja, o ponto em questão mais desafiador provavelmente é que seja desenvolvido um programa de MPT de forma endógena em um projeto mineral. / As it is directly under global economic factors, the mining industry tends to develop to seek leaner production processes, combining lower operational costs to better productive rates. In this context, the main objective of this study is to evaluate the general behavior of the time spend in preventive maintenance of a truck fleet from a large-scale mining project, being the analysis performed in a time window based on an economic period of local minimum in copper prices. From the observed behavior a discussion is made under the assumptions of the management methodology named Total Productive Maintenance in order to discuss the possible main risks and/or consequences that can be faced by the project and how these decisions qualitatively indicate the degree of maturity between production and maintenance in a mining project. In the end, it can be concluded that an intense work is still needed to translate the TPM premises into the mineral industry, being necessary to seek ways to adapt processes and metrics that are more suited to the extractive industry environment (which is very different from the manufacturing industry, where the program started). In addition, it is clear that the application of TPM is possible and welcome in mining since there are several academic studies in the area - in other words, the most challenging point in question is probably that an endogenous TPM program is developed within a mineral project.

Por estar diretamente sujeita a fatores econômicos de nível global, a mineração tende a se desenvolver para buscar processos produtivos mais enxutos, aliando menores custos operacionais a melhores taxas produtivas. Neste contexto, o objetivo principal deste trabalho é avaliar o comportamento geral da curva de tempo em manutenção preventiva de uma frota de caminhões em uma mineração de grande porte, sendo a análise realizada em uma janela temporal baseada em um período econômico de mínimo local nos preços do cobre. A partir do comportamento observado é feita uma discussão sob as premissas da metodologia de gestão nomeada Manutenção Produtiva Total de forma a se debater os possíveis principais riscos e/ou consequências que podem ser enfrentadas pelo projeto e como estas decisões qualitativamente indicam o grau de maturidade na integração operacional entre produção e manutenção em um projeto minerário. Ao final, pode-se concluir que ainda é necessário um intenso trabalho em se traduzir as premissas da MPT para a indústria mineral, buscando-se adaptar processos e métricas que sejam mais condizentes ao ambiente da indústria extrativa (que é muito diferente da indústria de manufatura, onde o programa teve início). Além disso, fica claro que a aplicação da MPT é possível e bem-vinda na mineração já havendo de vários estudos acadêmicos na área - ou seja, o ponto em questão mais desafiador provavelmente é que seja desenvolvido um programa de MPT de forma endógena em um projeto mineral. / As it is directly under global economic factors, the mining industry tends to develop to seek leaner production processes, combining lower operational costs to better productive rates. In this context, the main objective of this study is to evaluate the general behavior of the time spend in preventive maintenance of a truck fleet from a large-scale mining project, being the analysis performed in a time window based on an economic period of local minimum in copper prices. From the observed behavior a discussion is made under the assumptions of the management methodology named Total Productive Maintenance in order to discuss the possible main risks and/or consequences that can be faced by the project and how these decisions qualitatively indicate the degree of maturity between production and maintenance in a mining project. In the end, it can be concluded that an intense work is still needed to translate the TPM premises into the mineral industry, being necessary to seek ways to adapt processes and metrics that are more suited to the extractive industry environment (which is very different from the manufacturing industry, where the program started). In addition, it is clear that the application of TPM is possible and welcome in mining since there are several academic studies in the area - in other words, the most challenging point in question is probably that an endogenous TPM program is developed within a mineral project.

Syftet med denna uppsats är att undersöka en komponents    optimala utbyts" tid med hänsyn till kostnad och risk,    och föreslå schemalagda underhåll, med hjälp av statistiska metoder.    Genom att använda statistiska verktyg och historiska data, kan    en komponents samt systemets brister predikteras. När    forskaren vet hur ett system beter sig, kan dess fördelar    exploateras och tas till vara på. Schemaläggning av    förebyggande service, kostnads prognoser samt    uppskattning av förlängda garantier är möjliga fördelar som    kan nyttjas av denna rapport. Detta medför en högre    tillgänglighet och förbättrat rykte hos kund.    Tillförlitligthet teori är en viktigt del av    Total Quality Management, TQM, som säkerhetsställer kvalité.    I denna uppsats jämförs, förklaras och verifieras 2 kända ARP,    och strategin att ersätta endast då komponenten går sönder i en fallstudie.    Denna uppsats indikerar att en ARP med ändlig horisont ger mer optimalta resultat än    en ARP med oändlig horisont eller då ingen utbytespolicy används.    Barlow \& Proschan visade detta redan 1962.    I denna uppsats påvisas att ARP-teorier kan minska omkostnader och stilleståndstid    samt öka tillgängligheten. / The purpose of this thesis is to examine when a part's optimal   replacement time occurs in terms of risk and cost, and provide maintenance plans   accordingly using statistical methods.   With the use of statistical tools and historical data,   the failures of components as well as the system can be predicted.   Once the researcher knows how the system behaves, he/she can reveal the gains that   can be made. Scheduling of preventive maintenance, improved warranty cost   forecasts and estimation of lengthened warranty costs are   plausible benefits from this report. This will further result   in higher availability and improved reputation among clients.   Reliability theory is an important part of Total Quality Management (TQM),   ensuring good quality.   This thesis will compare the differences between two known age replacement policies (ARP),   and with the strategy of replacing only on failures in a real case-study.   This thesis indicates that an ARP with finite horizon yields   a more optimal solution than an ARP with infinite horizon as well as using no replacement policy at all.   Barlow & Proschan established this as far back as 1962.   With the aid of ARP theories it has been shown in this thesis that lowering costs is possible   and in the progress lower downtime which increases availability.

<p>Syftet med denna uppsats är att undersöka en komponents    optimala utbyts" tid med hänsyn till kostnad och risk,    och föreslå schemalagda underhåll, med hjälp av statistiska metoder.    Genom att använda statistiska verktyg och historiska data, kan    en komponents samt systemets brister predikteras. När    forskaren vet hur ett system beter sig, kan dess fördelar    exploateras och tas till vara på. Schemaläggning av    förebyggande service, kostnads prognoser samt    uppskattning av förlängda garantier är möjliga fördelar som    kan nyttjas av denna rapport. Detta medför en högre    tillgänglighet och förbättrat rykte hos kund.    Tillförlitligthet teori är en viktigt del av    Total Quality Management, TQM, som säkerhetsställer kvalité.    I denna uppsats jämförs, förklaras och verifieras 2 kända ARP,    och strategin att ersätta endast då komponenten går sönder i en fallstudie.    Denna uppsats indikerar att en ARP med ändlig horisont ger mer optimalta resultat än    en ARP med oändlig horisont eller då ingen utbytespolicy används.    Barlow \& Proschan visade detta redan 1962.    I denna uppsats påvisas att ARP-teorier kan minska omkostnader och stilleståndstid    samt öka tillgängligheten.</p> / <p>The purpose of this thesis is to examine when a part's optimal   replacement time occurs in terms of risk and cost, and provide maintenance plans   accordingly using statistical methods.   With the use of statistical tools and historical data,   the failures of components as well as the system can be predicted.   Once the researcher knows how the system behaves, he/she can reveal the gains that   can be made. Scheduling of preventive maintenance, improved warranty cost   forecasts and estimation of lengthened warranty costs are   plausible benefits from this report. This will further result   in higher availability and improved reputation among clients.   Reliability theory is an important part of Total Quality Management (TQM),   ensuring good quality.   This thesis will compare the differences between two known age replacement policies (ARP),   and with the strategy of replacing only on failures in a real case-study.   This thesis indicates that an ARP with finite horizon yields   a more optimal solution than an ARP with infinite horizon as well as using no replacement policy at all.   Barlow & Proschan established this as far back as 1962.   With the aid of ARP theories it has been shown in this thesis that lowering costs is possible   and in the progress lower downtime which increases availability.</p>

As a result of global competition, companies within the automotive industry are required to improve their production systems to achieve sustainable and competitive production. To attain operational excellence, companies have adopted concepts like lean and World Class Manufacturing (WCM). In order to deliver, companies need to increase the availability and reliability of production equipment. This demands focus on both operational and strategic maintenance, and efficient equipment acquisition and management. The main objective of this thesis is to identify processes and success factors that are necessary for improved maintenance performance. By conducting a case study and by performing a literature review, the following has been studied: 1) how companies within the automotive industry work with maintenance and maintenance strategies, 2) implementation of maintenance strategies, and 3) new acquisition and spare parts management for maintenance.  The results show that basic maintenance concepts and related terms are well defined in the literature. However, maintenance strategy is not unanimously defined. Results from the case study reveal that Professional Maintenance (PM) within WCM has been considered as a maintenance strategy. Furthermore, driving forces and obstacles for implementing maintenance strategies have been identified. The study also reveals that there is a strong link between maintenance performance, new acquisition and spare parts management. / Den globala konkurrensen inom fordonsindustrin, har medfört krav på företag att effektivisera sina produktionssystem för att åstadkomma en hållbar och konkurrenskraftig produktion. Detta har resulterat i arbete enligt koncept som lean och World Class Manufacturing (WCM).  För att kunna leverera behöver företag öka den tekniska tillgängligheten på utrustning. Detta i sin tur ställer ett ökat krav på både operativt och strategiskt arbete med underhåll, anskaffning och reservdelsstyrning. Målet med examensarbetet är att identifiera processer och faktorer som bidrar till ett förbättrat underhåll. Med hjälp av litteraturstudie och en fallstudie har följande studerats: 1) hur företag inom fordonsindustrin arbetar med underhåll och underhållsstrategier, 2) implementering av underhållsstrategier, och 3) nyanskaffning och reservdelsstyrning för underhåll. Resultat från litteraturstudier visar att det finns en enighet om att underhåll delas in i förebyggande och avhjälpande underhåll. Däremot, visar resultatet att det inte finns en enig definition av underhållsstrategi. Fallstudien visar att Professional Maintenance (PM) inom WCM betraktas som en underhållsstrategi. Som ett resultat har även ett antal drivkrafter och svårigheter som påverkar implementering av underhållsstrategier identifierats. Vidare visar resultatet att det finns ett starkt samband mellan underhåll, nyanskaffning och reservdelsstyrning.

Företag kan använda sig av olika strategier för underhåll. Det kan vara avhjälpande underhåll som utförs först när ett fel uppkommer eller förebyggande underhåll. Ofta används en kombination av båda. Det förebyggande underhållet kan vara förutbestämt eller tillståndsbaserat. Fördelar med att använda sig av tillståndsbaserat underhåll är exempelvis att utbyte av enheter inte behöver ske i onödan utan först när en observation visar att ett fel är på väg att ske. Vid val av förebyggande underhåll kan RCM-metoden användas för att utröna vilka feltillstånd som kan uppstå och som orsakar funktionsfel. Val av förebyggande underhåll kan baseras på fakta om nya tekniker, felhistorik samt vilken utrustning och kompetens som finns tillgänglig. Volvo Cars i Skövde använder sig just nu av i stort sett likadant förebyggande underhåll på nya verktygsmaskiner som på gamla. Företaget använder sig av tillståndsbaserat underhåll på vissa komponenter och enheter. Det består då främst av inspektioner, subjektiva tillståndskontroller, och vibrationsövervakning på motorspindlar. Genom intervjuer framkommer att nya tekniker sällan införs och att lagda planer sällan ändras. Företaget är dock medvetet om att nya tekniker för tillståndsbaserat underhåll finns och är intresserat av att undersöka om ett utökat tillståndsbaserat underhåll är lämpligt på verktygsmaskiner. De tillståndsbaserade teknikerna som det talats om är främst ballbarmätning, övervakning av processparametrar och geometrisk verifiering. Ett underlag för att se om nya tekniker bör införas togs fram genom datainsamling av felhistorik på maskiner. Avgränsning gjordes genom att en fem år gammal bearbetningslinje för cylinderhuvuden studerades. Genom sortering av data från haverier ur underhållssystemet Maximo från de senaste åren och prioritering genom Paretoprincipen visade det sig att återkommande haverier på kulskruvar finns. Till komponenten finns ett tillståndsbaserat underhåll kopplat, den inspekteras en gång om året. Det är dock sällan som fel på komponenten upptäcks på inspektionen, bara fyra av tjugosju bytta kulskruvar de senaste fem åren är bytta efter anmärkning från inspektion. Genom att kartlägga komponentens haverier på femton verktygsmaskiner och när FU gjorts hittades flera förbättringsmöjligheter. Kostnaden för att byta kulskruv när väl fel uppstått är upp till tio gånger så stor jämfört med att byta den förebyggande. När fel uppstått har det medfört kassationer av producerade detaljer och lång stilleståndstid. Motiv för att införa ett bättre förebyggande underhåll finns därför framför allt i minskade kostnader och ökad tillgänglighet. Företaget rekommenderas att införa ballbarmätning som ett nytt tillståndsbaserat underhåll. Detta för att öka den objektiva kontrollen av mekaniska slitdelar. Mätmetoden är inte helt okänd för företaget och relativt lätt att införa. Den subjektiva kontrollen som görs idag bör göras om för att bättre fånga begynnelse till fel. Kontrollen kräver idag en hel del driftstoppstid vilket gjort att den inte utförts i tid och bör därför kortas ner. Företaget rekommenderas också att undersöka nya metoder för förebyggande underhåll kontinuerligt. Idag finns ingen anställd med uppdraget och för att utveckla underhållet måste företaget hålla sig uppdaterade på nya tekniker. / A company can use different strategies when performing maintenance. One strategy is corrective maintenance where maintenance is carried out after a failure has occurred. Another strategy is preventive maintenance. A combination of the two is often used. The preventive maintenance can be predetermined or condition based. Advantages with the condition based maintenance are e.g. that the replacement of parts can be planned according to observations indicating there’s an increasing risk of failure. When choosing preventive maintenance, the RCM method can be used to determine risk of failure modes and the cause of function failures. Decisions regarding the preventive maintenance should be based on knowledge in new techniques, history of failures and the equipment and competence available. Volvo Cars in Skövde is doing more or less the same preventive maintenance on machine tools regardless of whether it’s a new or old machine. The company is using condition based maintenance on some of the components and units. Mostly it consists of human inspections, a subjective control of the machine’s condition, and spindle vibration analysis. Through interviews it has been determined that new techniques are rarely implemented and that the decided maintenance plan for a machine is rarely changed. The company is aware of the development of new techniques for condition based maintenance and wants to investigate whether it is appropriate to evolve their condition based maintenance on machine tools. The techniques further investigated are measurement with Ball bar, monitoring of process parameters and geometrical verification. To make accurate decisions regarding the implementation of new techniques, a data collection of the history of frequent failures was done. To limit the amount of data a five year old processing line was studied. Data from the maintenance system Maximo was sorted by long failures from the recent years and by prioritizing with the Pareto principle it showed the recurring failures of ball screws. The component do have a condition based maintenance activity today, there is an inspection once a year. However, the inspection is rarely able to predict or find a failure on the ball screw. Four out of twenty-seven ball screws with fault have been changed after inspection over the last five years. By mapping the components failures on fifteen machine tools and when preventive maintenance has been done, several options for improvements were found. The cost of the replacement of a ball screw when a failure has already occurred is as much as ten times the cost of replacement before failure. This is caused by cost of scrap and long down time. A new, improved preventive maintenance task is therefore highly justified. The motif is both reducing the scrap cost related to failure and with increased availability. The company is recommended an investment in Ball bar as a new method for condition based maintenance. It should be used to increase the objective control of the mechanical parts of a machine tool. The method is not totally new for the company and relatively easy to implement. The current subjective control on the machine tools should be revised to increase the probability of finding degradation of functions. Today the activity takes rather long time and has therefore not been done in time, it should be shorter. The company is also recommended to investigate new methods for preventive maintenance continually. Today there is no employee with that mission and to evolve maintenance, the company has to be updated on new techniques.