• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 29
  • 27
  • 19
  • 10
  • 3
  • 3
  • 1
  • 1
  • 1
  • Tagged with
  • 105
  • 105
  • 26
  • 24
  • 23
  • 23
  • 20
  • 18
  • 17
  • 16
  • 13
  • 11
  • 11
  • 11
  • 10
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
101

Erfarenhet mot Algoritmer : prediktivt underhåll genom maskininlärning i svenska Försvarsmakten / Experience versus Algorithms : Predictive Maintenance through Machine Learning in the Swedish Armed Forces

Ellmén, Niclas January 2024 (has links)
Rapporten utforskar möjligheten till implementering av prediktivt underhåll genom maskininlärning för markfordon i svenska Försvarsmakten. Undersökningen syftade till att finna nya metoder att planera och bedriva underhåll av militära fordon. Detta baserades på en litteraturstudie där civila och militära tillämpningar av maskininlärning granskades. Därefter analyserades empirin utifrån Kent Anderssons teori om militär nytta.  Analysen visar att det ur en teknologisk synpunkt skulle vara möjligt att implementera maskininlärning för prediktiva analyser av underhåll i militära fordon. Däremot tillkommer många utmaningar kopplade till datainsamling, interoperabilitet, cybersäkerhet, krav på höga kompetenser och ett behov av avancerade algoritmer. Militär verksamhet präglas även av mycket osäkerhet och fordon nyttjas i väldigt varierande miljöer. Detta försvårar omständigheterna för en AI att förutspå framtida utfall.  För att prediktiva analyser skulle kunna nyttjas effektivt skulle Försvarsmakten behöva ändra sitt förhållningssätt till materielunderhåll. Vidare forskning är emellertid nödvändig för att undersöka en faktisk implementering, eller ett maskininlärningssystems fullständiga militära nytta. / This work explores the possibility of implementing predictive maintenance through machine learning for ground vehicles in the Swedish Armed Forces. The investigation aimed to find new methods for planning and carrying out maintenance of military vehicles. This was based on a literature review that reviewed civilian and military applications of machine learning. The empirical data was then analyzed based on Kent Andersson's theory of military utility.  The analysis shows that from a technological point of view it would be possible to implement machine learning for predictive analysis of maintenance in military vehicles. However, there are many challenges linked to data collection, interoperability, cyber security, requirements for high skills and a need for advanced algorithms. Military operations are also characterized by a lot of uncertainty and vehicles are used in very varied environments. This makes it difficult for an AI to predict future outcomes.  For predictive analyzes to become a reality, the Swedish Armed Forces would have to change their approach to maintenance. However, further research is necessary to investigate an actual implementation and the full military utility of a machine learning system.
102

An improved maintenance management strategy for gas field equipment in Escravos gas–to–liquid plant, Nigeria / T.E. Onyenanu

Onyenanu, Tochukwu Emmanuel January 2010 (has links)
The safety record of most petrochemical industries in the world and Nigeria in particle, has not been able to come down to the maximum allowable range of 0 - 0.1 percent of tolerance on recordable injuries, due to increasing failure rates of equipment within the plant. Investigations on the maintenance audit carried out on the Nigerian Gas Company (NGC) revealed that 85 percent of such failures are directly linked to improper adaptability of an effective maintenance management strategy and plan within the petrochemical industries in Nigeria. Equally, the growth and continuous operation of any plant depends to a large extent on the maintenance of the equipment that refines the Crude Oil and natural Gas. As such, various maintenance management systems have been used over the years for the actualisation of the above purpose but with minimal success. This is evident in the fact that the level of maintenance performance of most Nigerian Petroleum Companies is always on the corrective maintenance model, which indirectly implies that the plant normally breaks down before maintenance management is applied. A critical look at the deficiency of improper adaptability of these maintenance management plans have conspicuously manifested in five major categories of maintenance failures which includes the following; * Failure of safety critical equipment due to lack of maintenance * Human error during maintenance * Static or spark discharge during maintenance in an intrinsically unsafe zone * Incompetence of maintenance staff, and * Poor communication between maintenance and production staff. These gaps as identified in this research must be corrected in the Nigerian Gas Industry if meaningful progress is to be made. Gas – To – Liquid technology is a very complex technology and with natural gas as the basic raw material, the technology not only looks intimidating but also is full of potential hazards. People are naturally afraid of the complex nature of gas in a confined environment (because of its highly combustible nature), its gaseous state makes it more complex for it to be kept under control and at the same time be moved from one form to another at different temperatures and pressures. The maintenance audit carried out on the Nigerian Gas Company (NGC) revealed some major loopholes in the maintenance management strategies adopted in the country. The audit reveals that the degree of adherence to conditions attached towards the maintenance management strategy of this equipment (in this case Gas field equipment) was too poor. Based on the above, this research is meant to improve the existing maintenance management strategy, by developing a Maintenance Management Strategy (MMS) that will be suitable for gas field equipment in the Escravos Gas–To– Liquid (EGTL) plant, planned to be commissioned in Nigeria early 2011. The need to research the above mentioned Maintenance Strategy became imperative due to the fact that the rate at which most of the petrochemical plants in the world are being gutted by fire, mainly due to poor maintenance management systems is alarming. This research work proffered solutions that will reduce or completely eliminate the highlighted problems above. This was based on investigations and analysis carried out in the chosen research area. Models were developed for the actualization of this Improved Maintenance Management Strategy (IMMS), so that the desired safe operability of the gas field equipment in the Escravos Gas–To– Liquid (EGTL) plant will be achieved without maintenance failure of any kind. / Thesis (M.Ing. (Development and Management Engineering))--North-West University, Potchefstroom Campus, 2011.
103

An improved maintenance management strategy for gas field equipment in Escravos gas–to–liquid plant, Nigeria / T.E. Onyenanu

Onyenanu, Tochukwu Emmanuel January 2010 (has links)
The safety record of most petrochemical industries in the world and Nigeria in particle, has not been able to come down to the maximum allowable range of 0 - 0.1 percent of tolerance on recordable injuries, due to increasing failure rates of equipment within the plant. Investigations on the maintenance audit carried out on the Nigerian Gas Company (NGC) revealed that 85 percent of such failures are directly linked to improper adaptability of an effective maintenance management strategy and plan within the petrochemical industries in Nigeria. Equally, the growth and continuous operation of any plant depends to a large extent on the maintenance of the equipment that refines the Crude Oil and natural Gas. As such, various maintenance management systems have been used over the years for the actualisation of the above purpose but with minimal success. This is evident in the fact that the level of maintenance performance of most Nigerian Petroleum Companies is always on the corrective maintenance model, which indirectly implies that the plant normally breaks down before maintenance management is applied. A critical look at the deficiency of improper adaptability of these maintenance management plans have conspicuously manifested in five major categories of maintenance failures which includes the following; * Failure of safety critical equipment due to lack of maintenance * Human error during maintenance * Static or spark discharge during maintenance in an intrinsically unsafe zone * Incompetence of maintenance staff, and * Poor communication between maintenance and production staff. These gaps as identified in this research must be corrected in the Nigerian Gas Industry if meaningful progress is to be made. Gas – To – Liquid technology is a very complex technology and with natural gas as the basic raw material, the technology not only looks intimidating but also is full of potential hazards. People are naturally afraid of the complex nature of gas in a confined environment (because of its highly combustible nature), its gaseous state makes it more complex for it to be kept under control and at the same time be moved from one form to another at different temperatures and pressures. The maintenance audit carried out on the Nigerian Gas Company (NGC) revealed some major loopholes in the maintenance management strategies adopted in the country. The audit reveals that the degree of adherence to conditions attached towards the maintenance management strategy of this equipment (in this case Gas field equipment) was too poor. Based on the above, this research is meant to improve the existing maintenance management strategy, by developing a Maintenance Management Strategy (MMS) that will be suitable for gas field equipment in the Escravos Gas–To– Liquid (EGTL) plant, planned to be commissioned in Nigeria early 2011. The need to research the above mentioned Maintenance Strategy became imperative due to the fact that the rate at which most of the petrochemical plants in the world are being gutted by fire, mainly due to poor maintenance management systems is alarming. This research work proffered solutions that will reduce or completely eliminate the highlighted problems above. This was based on investigations and analysis carried out in the chosen research area. Models were developed for the actualization of this Improved Maintenance Management Strategy (IMMS), so that the desired safe operability of the gas field equipment in the Escravos Gas–To– Liquid (EGTL) plant will be achieved without maintenance failure of any kind. / Thesis (M.Ing. (Development and Management Engineering))--North-West University, Potchefstroom Campus, 2011.
104

The environmental and social impact from digitization in buildings : A case study of the transformation and current conditions on the University hospital of Northern Sweden

Melén, Matilda, Wenhov, Alma January 2022 (has links)
To improve sustainability, social, environmental and economic aspects needs to be considered. The most optimal result appears when all three aspects are balanced equally, this is however often overseen by private investors, who focuses only on reaching economic sustainability at the expense of social and environmental sustainability. Building digitization is one way to potentially improve the sustainability of a building socially, environmentally and economically. Focusing on the aspects that often are neglected, this thesis aims to investigate if digitizing a building improves social and environmental sustainability. The investigation is made by evaluating the implementation of a digital building automation tool from selected social and environmental sustainability criteria at the University hospital of Northern Sweden. This by performing an interview survey with the maintenance organisation and the tenants in the building, as well as performing CO2-e calculations on emissions connected to energy usage, transportation and production of HVAC-products. The evaluation indicated that the implementation had resulted in improved sustainability in the studied building, both socially and environmentally. Showing that digitizing a building improves social and environmental sustainability. The social sustainability had been positively affected from increased efficiency and effectiveness of the maintenance work in the building and improved well-being of the maintenance staff. However, the tenants were not completely satisfied when asked if reported errors were being solved, but on the other hand had the maintenance organisation experienced an improvement in satisfaction among the tenants since the implementation of the digital building automation tool. Furthermore, the tenants were generally more satisfied than dissatisfied with the indoor climate, except for experienced low temperatures in the winter and dry air. The environmental sustainability had been improved from a reduction in emitted CO2-e, generated from less energy usage and minimized transportation connected to maintenance operations. Furthermore, an estimation on increased technical lifetime of HVAC-products demonstrated on a potential further reduction in emitted CO2-e during the building's whole life span. Finally, the evaluation identified two combined effects between social and environmental sustainability. First, the increased efficiency and effectiveness of the maintenance work was one direct factor for the decrease in CO2-e emissions connected to transportation. Second, the tenants in the building expressed that they would feel prouder of their choice of employer if their employer focused more on reducing their climate impact, which motivates to work with environmental sustainability to achieve satisfaction and moreover an improved social sustainability.  The results of this case study indicate that digital building automation improves the social and environmental sustainability of a building, strengthening the statement on potential sustainability improvements from building digitization. For property owners wanting to increase their building's sustainability, digital building automation is therefore a proposed course of action. However, the performance on sustainability and the balance between the different aspects should continuously be evaluated, as the study showed that further improvements on the social and environmental sustainability still could be made. Property owners working towards an improved sustainability through digitization will both see long term positive effects on people's health, as well as help fulfilling the climate goals in the Paris agreement, resulting in a more sustainable world for present and future generations. / För att skapa en bättre hållbarhet måste hänsyn tas till både den sociala, miljömässiga samt den ekonomiska aspekten. Fokuset på dessa tre aspekter bör balanseras lika för att uppnå optimal hållbarhet, dock efterföljs detta inte vanligtvis av privata investerare som ofta endast fokuserar på att uppnå ekonomisk hållbarhet på bekostnad av den sociala- och miljömässiga hållbarheten. En byggnads hållbarhet kan potentiellt förbättras genom digitalisering av byggnaden, vilket kan förbättra den sociala, miljömässiga och ekonomiska hållbarheten. Genom att fokusera på de aspekter som ofta blir nedprioriterade undersöker denna studie om den sociala och miljömässiga hållbarheten i en byggnad ökar vid digitalisering. Detta genom att utvärdera implementeringen av ett digitalt verktyg på Norrlands Universitetssjukhus från några utvalda kriterier för social och miljömässig hållbarhet. Den sociala hållbarheten utvärderas med hjälp av en intervjustudie med underhållspersonalen och hyresgästerna i byggnaden, medan den miljömässiga hållbarheten utvärderades genom att beräkna CO2-e utsläpp kopplade till energianvändning, transporter samt produktionen av HVAC-produkter. Utvärderingen indikerade att implementeringen hade resulterat i en ökad social och miljömässig hållbarhet i byggnaden, vilket visar att en byggnads sociala och miljömässiga hållbarhet kan förbättras genom digitalisering. Den sociala hållbarheten hade ökat på grund av en ökad effektivitet i underhållsarbetet samt ett förbättrat välmående hos underhållspersonalen. Dock visade undersökningen att hyresgästerna i byggnaden inte var helt nöjda med arbetet kring felanmälningar, underhållsorganisationen hade dock upplevt att hyresgästerna var nöjdare efter implementeringen av det digitala verktyget än de var innan. Hyresgästerna var också mer nöjda än missnöjda med inomhusklimatet, förutom att de upplevde låg inomhustemperatur på vintern och att luften var torr. Den miljömässiga hållbarheten hade förbättrats genom en minskning av CO2-e utsläpp från minskad energianvändning och minskade transporter kopplade till underhållsarbetet. Den estimerade ökningen av den tekniska livslängden på HVAC-produkter visade också på potentiella minskningar av CO2-utsläpp under byggnadens hela livslängd. Denna studie identifierade också två kombinerade effekter mellan social- och miljömässig hållbarhet. Den första var att genom ökad effektivitet i arbetet för underhållspersonalen så minskade även CO2-e utsläppen från transporterna. Den andra var att hyresgästerna i byggnaden uttryckte att de skulle känna sig stoltare över sitt val av arbetsgivare om deras arbetsgivare fokuserade mer på att minska sin klimatpåverkan, vilket motiverar att arbeta med miljömässighållbarhet för att uppnå ökad tillfredsställelse hos hyresgästerna och därav ökad social hållbarhet.  Resultaten från denna fallstudie indikerar att digital fastighetsautomation förbättrar den sociala och miljömässiga hållbarheten av en byggnad, vilket styrker argumentet att en byggnads hållbarhet potentiellt kan förbättras genom digitalisering. För fastighetsägare som vill öka en byggnads hållbarhet är digital fastighetsautomation därför rekommenderat. Hållbarheten samt balansen mellan de olika hållbarhetsaspekterna behöver däremot kontinuerligt utvärderas, eftersom denna studie har visat att den sociala- och miljömässigahållbarheten fortfarande kan förbättras. Fastighetsägare som arbetar med att öka hållbarheten i byggnader genom digitalisering kommer uppleva långvariga positiva effekter för människors hälsa samt så kommer de bidra till att uppnå klimatmålen i Parisavtalet. Detta resulterar i en mer hållbar värld för nutida och framtida generationer.
105

Fallstudie om Prediktivt och Tillståndsbaserat Underhåll inom Läkemedelsindustrin / Case study regarding Predictive and Condition-based Maintenance in the Pharmaceutical Industry

Redzovic, Numan, Malki, Anton January 2022 (has links)
Underhåll är en aktivitet som varje produktion vill undvika så mycket som möjligt på grund av kostnaderna och tiden som anknyts till den. Trots detta så är en väl fungerande underhållsverksamhet väsentlig för att främja produktionens funktionssäkerhet och tillgänglighet att tillverka. En effektiv underhållsorganisation går däremot inte ut på att genomföra mer underhåll än vad som egentligen är nödvändigt utan att genomföra underhåll i rätt tid. På traditionellt sätt så genomförs detta genom att ersätta slitage delar och serva utrustningen med fastställda mellanrum för att förebygga att haveri, vilket kallas för förebyggande underhåll. De tidsintervaller som angivits för service bestäms av leverantörerna och grundar sig i en generell uppskattning av slitagedelarnas livslängd utifrån tester och analys. Till skillnad från att köra utrustningen till den går sönder som kallas för Avhjälpande underhåll så kan underhåll genomföras vid lämpliga tider så att det inte påverkar produktion och tillgänglighet. Men de tidsintervall som leverantörerna rekommenderar till företagen garanterar inte att slitage delen håller sig till det intervallet, delarna kan exempelvis rasa tidigare än angivet eller till och med hålla längre. Av denna anledning är det naturliga steget i underhållets utveckling att kunna övervaka utrustningens hälsa i hopp om att kunna förutspå när och varför ett haveri ska uppstå. Den här typen av underhåll kallas för tillståndsbaserat och prediktivt underhåll och medför ultimat tillgänglighet av utrustning och den mest kostnadseffektiva underhållsorganisationen, då god framförhållning och översikt uppnås för att enbart genomföra underhåll när det behövs. Det som gör tillståndsbaserat och prediktivt underhåll möjligt är den fjärde industriella revolutionen “Industri 4.0” och teknologierna som associeras med den som går ut på absolut digitalisering av produktionen och smarta fabriker. Teknologier som IoT, Big Dataanalys och Artificiell Intelligens används för att koppla upp utrustning till nätet med hjälp av givare för att samla in och lagra data som ska användas i analyser för att prognosera dess livslängd. Uppdragsgivaren AstraZeneca i Södertälje tillverkar olika typer av läkemedel som många är livsviktiga för de patienter som tar dessa mediciner. Om AstraZenecas produktion står still på grund av fel i utrustningen kommer det inte enbart medföra stora ekonomiska konsekvenser utan även påverka de människor som med livet förlitar sig på den medicin som levereras. För att försäkra produktionens tillgänglighet har AstraZeneca gjort försök att tillämpa tillståndsbaserat och prediktivt underhåll men det är fortfarande enbart i startgroparna. Eftersom ventilation är kritisk del av AstraZeneca produktion då ett fel i ventilationssystemet resulterar i totalt produktionsstopp i byggnaden förens problemet åtgärdas och anläggningen sanerats blev det även rapportens fokusområde. Arbetets uppgift går därför ut på att undersöka möjligheter för AstraZeneca att utveckla deras prediktiva och tillståndsbaserat underhåll på deras ventilationssystem, för att sedan kunna identifiera och presentera förslag på åtgärder. Dessa förslag analyserades sedan med hjälp av verktygen QFD-Matris och Pugh-Matris för att kunna uppskatta vilket förslag som är mest kostnadseffektivt, funktions effektivt samt vilket förslag som kommer tillföra mest nytta för underhållet på AstraZeneca. / Maintenance is an activity that every production wants to avoid as much as possible due to the costs and the time associated with it. Despite this, a well-functioning maintenance operation is essential to promote the production's availability to manufacture and operational reliability. Running an efficient maintenance operation is not about carrying out more maintenance than is necessary but carrying out the right amount of maintenance at the right time. Traditionally speaking this is done by replacing worn parts and servicing the equipment at set intervals to prevent breakdowns, this method is called preventive maintenance. The intervals specified for service are determined by the suppliers and are based on general estimates of the service life for the spare parts from test and analytics. Preventive maintenance allows for maintenance to be carried out at appropriate time to not affect production and availability unlike running the equipment until breakdown, which is called reactive maintenance. However, these intervals that the suppliers recommend do not guarantee that the parts adhere to the given interval, the part can for example break down earlier than expected or even outlast its prescribed lifetime. Because of this, the natural step in the development of maintenance is giving companies the ability to monitor the health of the equipment in hope of being able to predict potential breakdowns. This is what Condition-Based and predictive maintenance is and it provides the ultimate availability of equipment and the most cost-effective maintenance organization, because the good foresight and overview allows maintenance to be carried out only when needed. The fourth industrial revolution “Industry 4.0”, absolute digitalization of production, smart factories and all the technologies associated with this is what makes this type of maintenance possible. Technologies such as IoT, Big Data Analytics and Artificial Intelligence are used to connect equipment to the network using sensors so that data can be stored and collected to be analyzed to forecast the lifespan of parts and equipment. AstraZeneca in Södertälje manufactures different types of medicine, many of which are vital for the patients who take them. If their production comes to a standstill due to equipment failure, it will not only have major financial consequences but also greatly affect the people who rely on the medicine offered with their lives. To ensure the availability of production, AstraZeneca has made attempts to apply condition-based and predictive maintenance, but it is still only in its infancy. Since ventilation is a critical part of AstraZeneca's production, as a failure here will result in a total production stoppage for the building affected and will not resume before the problem is remedied and the plant is decontaminated, it also became the report's focus area. The task at hand is therefore to investigate the opportunities AstraZeneca must develop their predictive and condition-based maintenance for their ventilation systems, in order to be able to present proposals for measures. The proposals will then be analyzed using tools like the QFD-Matrix and the Pugh-Matrix in order to estimate which is more cost effective, function effective and which one will bring the most benefit to AstraZeneca.

Page generated in 0.078 seconds