• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 9
  • Tagged with
  • 9
  • 9
  • 9
  • 6
  • 3
  • 3
  • 3
  • 3
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 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.
1

Development and Evaluation of Alternative Electric Thermostat Design

Isaksson, Arvid January 2018 (has links)
Introducing an electric thermostat to the thermal management system is a way of actively controlling the temperature of the engine, which has been shown to have several possible gains regarding power, fuel consumption, emissions and engine durability. Complexity, cost and durability are key concerns that have led to no heavy duty truck on the market having an electrically controllable thermostat. This emphasizes the need for exploring alternative solutions that enables electric control of the thermostat according to the needs of heavy commercial vehicles. Several concepts have been generated to solve this problem and a model based approach in Simulink, Matlab and GT Suite was used for the development and evaluation. The most promising concept of combining a BLDC electric motor with a wax body enables electric control with a downsized actuator and full fail-safe function while showing improvements in temperature control performance compared to a traditional wax thermostat. This thesis has increased the knowledge on the subject and could allow for implementing an electrically controlled thermostat in future Scania heavy duty trucks, leading to a more durable engine with lower fuel consumption and emissions.
2

Production as a Source of Competitive Advantage in the Value Chain of Electrified Heavy Vehicles Manufacturing Companies

Nabati, Hamid, Banaei Fard, Mohsen January 2021 (has links)
Global goals to reduce the greenhouse gases and CO2 emission has challenged many industries to innovate new technologies or shift to different products. Electrification of transportation has been considered a major improvement in this respect, which affects the entire transportation value chain including the manufacturing of heavy-duty vehicles. The importance and degree of changes that shifting toward electrification imposes on the truck manufactures value chain is subject to the study with a focus on the production stage of the value chain. The research also creates some insights to understand the sources of competitive advantage in the production of BEV.    Based on the authors’ knowledge and investigations, the contribution of this study in the research field is that the concept of the value chain together with Delphi method is implemented for the first time in the BEV industry to identify the impact of technology transformation on the production activities and competitive advantage. Furthermore, it contributes to highlight the unclarities in the BEV value chain activities that need more researches in order to create a more clear technology transformation path for the heavy-duty vehicle industry. The research findings imply that electrification has a big impact on the production value chain activities. The number of employees will be reduced that in turn creates a source of cost advantage for the company by reducing the cost per vehicle production. The assembly time is also shorter that makes it easier to respond to the market dynamic faster and reduces the cost of products storage. However, the risk for production stops is considered higher in BEV compare to ICE vehicles. Another finding is that the outsourcing of the components has a negative effect on the competitive advantage of the company and production efficiency. There is no difference between the impact of production activities on the company value chain in both BEV and ICE manufacturing. There is also a consensus on the more opportunities in BEV manufacturing to increase the production efficiency and lowering the cost.
3

The environmental impact assessment of manual and automatic truck types

Maithani, Abhinav January 2020 (has links)
Trucks are used to transport goods and services using road transportation. Emissions from the transportation industry are rising with every passing year. The Swedish government said in a report in 2009 that they are working to reduce 70% of the greenhouse gas emissions by 2030 to build a pathway for sustainable transportation (Allerup, 2019). New technologies in the field of transportation are coming to the market every year. It is important to evaluate the impacts from these technologies to support the 2030 agenda of Swedish government. Life cycle analysis of trucks can give valuable insight into the hotspots from different life cycle stages. To provide a comprehensive overview of the emissions from different types of heavy duty trucks namely Internal combustion engine trucks (Manual and driverless) and electric trucks (manual and driverless) a life cycle assessment is carried out.  From the comprehensive evaluation of results, following questions are answered. [1] Global warming potential from different types of heavy duty trucks; [2] Impacts from the new automation technology on climate change; [3] Environmental hotspots evaluation of the different truck types; and [4] Impacts of using different fuel types on the Global warming potential. Overall, the life cycle assessment can help the decision makers for taking decisions in order to fulfill the agenda 2030 of the Swedish government but still further research is needed to be done to exactly find out the emissions level of new automation technologies. / Lastbilar används för att transportera varor och tjänster med vägtransport. Utsläppen från transportbranschen ökar för varje år som går. Den svenska regeringen sa i en rapport 2009 att de arbetar för att minska 70% av växthusgasutsläppen till 2030 för att bygga en väg för hållbara transporter. Ny teknik inom transportområdet kommer till marknaden varje år. Det är viktigt att utvärdera effekterna av denna teknik för att stödja 2030-agendan för den svenska regeringen. Livscykelanalys av lastbilar kan ge värdefull inblick i hotspots från olika livscykelsteg. För att ge en omfattande översikt över utsläppen från olika typer av tunga lastbilar, nämligen IC-motor bilar (manuella och förarlösa) och elektriska lastbilar (manuella och förarlösa) genomförs en livscykelbedömning.  Från den omfattande utvärderingen av resultaten besvaras följande frågor. [1] Global uppvärmningspotential från olika typer av tunga lastbilar; [2] Effekter från den nya automatiseringstekniker på klimatförändringar. [3] Miljö Hotspots utvärdering av olika lastbilstyper; och [4] Effekter av användning av olika bränsletyper på den globala uppvärmning potentialen. Sammantaget kan livscykelbedömningen hjälpa beslutsfattarna att fatta beslut för att uppfylla den svenska regeringens dagordning 2030, men ytterligare forskning behövs för att exakt ta reda på utsläppsnivån för ny automatiseringsteknik.  Resultaten av studien visar att elbilen har lägre GWP-utsläpp än IC-motor bilarna. Den förarlösa automatiseringstekniker har 1,37% av de totala GWP-utsläppen när det gäller IC-motor bilar. Medan för elektriska lastbilar bidrar förarlös automatiseringsteknik upp till 12% av de totala utsläppen av elbil.
4

Development of a Method to Prepare a Test Fluid with Soft Particles in Biodiesel Fuel Blends / Utveckling av en metod för att framställa en testvätska av mjuka partiklar i bränsleblandningar med biodiesel

Jadwani, Vishal January 2020 (has links)
Bilindustrin är en av de största bidragsgivarna till koldioxidutsläpp genom fordonsutsläppnär de används. För att göra transportsektorn mer hållbar EU-politik dikterar att inkludera enhögre andel förnybara energikällor i den totala energiförbrukningen och ett sätt att uppnå det äratt införliva biobränslen i de traditionella fossila bränslena. Detta skulle minskakoldioxidavtrycket som genereras av bilindustrin på ett ekonomiskt sätt utan att göra storatekniska modifieringar i befintliga motorer. Scania AB har optimerat sitt bränslesystem för attuppfylla 10% blandning av biodiesel i normalt dieselbränsle.Scania AB är en lastbiltillverkare i Sverige med deras forsknings- och utvecklingscenterbeläget i Södertälje. De söker en lösning för att effektivisera sina bränslefilter i lastbilsmotorerför att filtrera bort mjuka partiklar som genereras på grund av nedbrytning av biobränsle. Syftetmed detta projekt kan delas upp i två huvudfaser; den första fasen är att utveckla en metod för attsimulera bränsleåldring och nedbrytning av biodiesel vilket leder till bildning av mjuka partiklarvid laboratorieförhållanden. Den senare fasen skulle vara att använda resultat från den förstafasen för att generera ett håravbrutet bränsle som därefter skulle användas för testning i enfullständig filtreringsrigg för att bedöma effektiviteten hos Scania-bränslefilter. Dessa testerskulle ge en inblick i befintliga filterprestanda beträffande nedbrutet bränsle och därmed kundebättre filter utformas för att effektivt hantera mjuka partiklar.I detta avhandlingsprojekt diskuteras olika iterationer och metoder för bildning av mjukapartiklar. Ett viktigt antagande baserat på verkliga deponeringsformationer från lastbilar överhela världen är att kalcium är en av de viktigaste orsakerna till bildning av mjuka partiklar.Resultat och bevis från tidigare avhandlingsprojekt har använts, modifierats och utvidgats i dettaprojekt. Nya metoder har utvecklats baserat på empiriska bevis från experimenten somgenomfördes under detta projekt som skulle förbättras ytterligare när Scania fortskrider meddetta projekt. Ett huvudkrav för ett håligt nedbrutet bränsle var att göra det stabilt när det gällersuspension av mjuka partiklar i biodieseln utan användning av konstant omröring för att simuleraförhållandena för bränsletanken i en lastbil.Resultaten som presenteras i detta avhandlingsprojekt är den optimerade metoden förframgångsrikt att producera mjuka partiklar och en arbetsmetod för att förberedatestbränslekoncentrat. Analys har utförts på testbränslekoncentrat i detta projekt för attkontrollera testfluidens livskraft för att utföra experiment på filtreringsrigg. Det har konstateratsatt tiofaldigt utspädningstestbränsle är det mest lovande provbränsleprovet som kan framställasmed de givna villkoren och tidsbegränsningarna. / The automobile industry is one of the largest contributors to carbon emissions through vehicularemissions when in use. To make the transport sector more sustainable EU policies dictate toinclude a higher share of renewables in total energy consumption and one way to achieve that isto incorporate biofuels in the traditional fossil fuels. This would reduce carbon footprintgenerated by the automobile industry in an economical manner without making majortechnological modifications in existing engines. Scania AB has optimized their truck fuel systemto comply with the 10% blend of biodiesel in normal diesel fuel.Scania AB is a truck manufacturer in Sweden, with their research and development centresituated in Södertälje. They seek a solution to make their fuel filters in truck engines efficient tofilter out soft particles generated due to biofuel degradation. The aim of this project can bedivided in two major phases; the first phase is to develop a method to simulate fuel ageing anddegradation of biodiesel leading to formation of soft particles at laboratory conditions. The laterphase would be to use results from the first phase to generate a mock degraded fuel that would besubsequently used for testing in a full size filtration rig to assess the efficacy of Scania fuelfilters. These tests would give an insight about existing filters performance regarding degradedfuel and thus better filters could be designed to efficiently handle soft particles.In this thesis project, different iterations and methods of formation of soft particles arediscussed. An important assumption based on real deposit formations from trucks across theworld is that calcium is one of major causes of soft particles formation. Results and evidencefrom previous thesis projects have been used, modified and extended in this project. Newmethods have been developed based on empirical evidence from the experiments conductedduring this project which would be further improved as Scania progresses with this project. Amajor requirement for a mock degraded fuel was to make it stable in terms of suspension of softparticles in the biodiesel without a use of constant agitation in order to simulate the conditions ofthe fuel tank in a truck.The results presented in this thesis project are the optimized method to successfullyproduce soft particles and a working method to prepare test fuel concentrate. Analysis has beenperformed on test fuel concentrate in this project to check for viability of test fluid forconducting experiments on filtration rig. It has been concluded that 10 folds dilution test fuel isthe most promising test fuel sample that can be prepared with the given conditions and timerestrictions
5

On Electrification of Heavy-Duty Trucks : A Grid Impact Analysis of Grid Integration of a High-Power Charging Station

Arvidsson, Maria January 2022 (has links)
The Swedish transport sector will need to undergo a major restructuring to achieve the established climate and environmental goals. The biggest change is that fossil fuels will be phased out and a larger part of the vehicle fleet will be electrified. This study deals with the electrification of heavy-duty trucks and how high-power charging stations affect the local electricity grid. Charging of heavy-duty trucks depends largely on the logistics of the transport system, which reduces the demand flexibility of power. High-power charging entails a risk of increased power peaks, which can affect the bus-voltage profiles, losses and loads on grid components. This thesis has been conducted as general study based on the case with the high-power charging station at Vädermotet in the area Hisingen of Gothenburg. The purpose was to build a generic model of the electricity grid at Hisingen and then investigate the consequences of high-power charging for the grid for two charging scenarios: the first scenario with four ABB Terra 360 chargers, and the second scenario with six ABB Terra 360 chargers and one MCS. The electricity network model and simulations were performed in PSS®SINCAL. The two charging scenarios, as well as the scenario before chargers were installed, were then simulated for three different system-load cases: maximum, average, and low load. The results showed that high-power charging of trucks had the biggest impacts for the voltage profiles during the case of low load. For the medium load and maximum load cases, the effect of the high-power charge decreased. Furthermore, electricity network losses increased for the low load case, but decreased slightly for the average and maximum load case. The reason was a more even load balance between the bus that connected the charging station to the grid and the rest of the network for the average and maximum load cases. In summary, the study indicated that grid implementation of a high-power charging station will have consequences for the local power system. However, the magnitude of the effects is not validated and can therefore only be regarded as indications. The outcome can be partly explained by the assumptions and simplifications of the model compared to the real system.
6

Strategic management of intellectual property in the field of vehicle autonomy and connectivity : A case study in the truck manufacturing field

Åkerblom, Philip January 2021 (has links)
The truck manufacturing field is expected to go through a disruptive change in the coming decade(s) due to a predicted implementation of autonomous (self-driving) heavy-duty vehicles. Here, companies face a challenge in that autonomous vehicles (AV) will include technologies and competences from multiple industries. This will most likely have an impact on how companies manage their intellectual property (IP) to protect their inventions if they wish to keep up with the changing market. To do this in the most optimal way, it is argued that a company’s IP-strategy needs to be incorporated into their business strategy. At the same time, studies show that companies are experiencing difficulties in extracting the full potential of their IP-management. Hence, this master thesis aims at investigating how a truck manufacturing company can adapt their IP-management policy in line with future developments within these emerging technologies and how their IP-strategy can become intertwined with their business strategy. Next to a literature review of these fields, the thesis conducts a qualitative case study of a truck manufacturer that intends to be part of this change based on 11 semi-structured interviews. From the empirical results, the company’s decision-making process regarding IP is identified, whereby this thesis suggests four points of improvement that could make their IP-management policy more effective. Further, this study presents three actions that can be taken in a company to incorporate their IP-strategy with their overall business strategy. The study ends with giving suggestions for future research and presenting six managerial implications that would come by implementing the suggestions made in this thesis.
7

High temperature corrosion in exhaust application for heavy-duty trucks

Moya, Alice January 2019 (has links)
Increasing awareness of environmental protection has made both governments and the industry aim for lowering carbon dioxide emissions. For the transport industry this means increasing engine efficiency, replacing fossil fuels with bio-based fuels or full electrification. For heavy-duty trucks, the first two options are currently the paths taken as short and mid-term solutions. These alternatives introduce new service conditions to the engines; namely higher combustion temperature and pressure, which will impose increased thermal and mechanical loads on the engine parts. In particular, the exhaust system parts must withstand constant thermal cycles in their normal operation. In heavy-duty trucks, exhaust systems are mostly manufactured from cast iron or cast steel. The usual materials, such as cast iron SiMo51 are reaching their maximum operating temperature, therefore new materials must come forward to fulfill new challenges. These materials oxidize at high temperatures forming different types of scales, which sometimes can act as protective barriers preventing their degradation. However, thermal cycles in the engine can impose new stresses and strains in these newly formed oxide scales, sometimes leading to spallation. A continuous spallation behavior in the exhaust systems is deleterious for the system, and the debris could also affect the downstream engine parts. This investigation focuses on the study of high temperature oxidation behavior of four iron-based candidate materials. Samples of two ductile cast irons and two austenitic stainless steels were exposed to thermal cycling in a simulated exhaust gas atmosphere at 850 °C, and to isothermal experiments at 850 °C and 900 °C in a stagnant air atmosphere. Additionally, the thermodynamics and kinetics were simulated using Thermo-Calc and Dictra software, respectively.  The results show that SiMo1000 grows a relatively thick iron-rich oxide layer with evidence of internal oxidation aided by the graphite shape exhibited by the alloy. The other cast iron Ni-Resist behaves better than SiMo1000, forming chromia and silica layers that prevent internal oxidation from occurring, although some spallation did occur in water containing atmospheres. 1.4832 behaved poorly compared to the other materials, entering into breakaway oxidation mode throughout all the exposures; therefore, it is not a material suited for high temperature service. HK30 was susceptible to water aided chromium evaporation but had a comparatively small mass change throughout the experiments; nonetheless, there was evidence of internal oxidation following interdendritic zones. Also casting defects were observed in these areas. Both might affect mechanical properties at high temperature. / Ökad miljömedvetenhet har gjort att både industri och politiker har satt upp mål för att sänka koldioxidutsläppen. För transportbranschen innebär detta till exempel ökad motoreffektivitet, att fossila bränslen ersätts med biobaserade bränslen eller full elektrifiering. För tunga lastbilar är de två första alternativen de som är aktuella på kort och medellång sikt. Dessa alternativ innebär nya förhållanden för motorn; nämligen högre förbränningstemperatur och -tryck, vilket kommer att öka termisk och mekanisk last på motorn. Detta är särskilt tydligt för avgassystemet, eftersom det utsätts för termisk cykling vid normal drift. Avgassystemet i en tung lastbil är oftast gjort i gjutjärn och ibland i gjutstål. Gjutjärnen, t ex segjärn SiMo51, börjar nu närma sig sin maximala driftstemperatur och nya material behöver introduceras. Vid höga temperaturer oxiderar dessa metalliska material och bildar olika typer av oxidskal. Beroende på sammansättningen på oxidskalet, kan det fungera som skydd för underliggande material. Termisk cykling kan ge spänningar i oxidskiktet som i sin tur kan ge flagning av skiktet. Om flagningen fortsätter kontinuerligt, förbrukas dels material, men flagorna kan också ge skador nedströms i avgassystemet. I detta arbete undersöks fyra järnbaserade kandidatmaterial avseende högtemperaturkorrosion. Prover av två gjutjärn och två austenitiska rostfria gjutstål exponerades dels isotermt vid 850 °C och 900 °C i stillastående luft, dels i experiment med termisk cykling i en simulerad flödande avgasatmosfär och varm temperatur 850 °C. Dessutom användes termodynamisk programvara (Thermo-Calc/DICTRA) för att simulera termodynamik och kinetik. Resultaten visar att SiMo1000 bildar ett relativt tjockt, järnrikt oxidskikt med viss inre oxidation som verkar följa grafitstråk i materialet. Det andra gjutjärnet, segjärnet Ni-Resist, beter sig bättre än SiMo1000 och bildar krom och kiseldioxidlager som förhindrar intern oxidation. Viss flagning observerades i den cykliska exponeringen. 1.4832 visade ett sämre beteende än de andra materialen och bildade inget skyddande oxidskikt, utan visade kontinuerlig massförlust i samtliga exponeringar. Detta material är därför inte lämpligt för de undersökta högtemperaturmiljöerna. HK30 visade låg massförlust i samtliga undersökningar med oxidation i interdendritiska områden. Även gjutfel som t ex porer observerades idessa områden. Båda kan påverka materialets mekaniska egenskaper vid dessa temperaturer.
8

Truck electrification : Trends and impacts on the energy system

NORDHAMMER, CAROLINE, GRANKVIST, JESSICA January 2021 (has links)
In line with the Paris agreement, Sweden has set up a national emission reduction target for the transport sector; to reduce the emissions by 70% by 2030, relative to levels in 2010. This entails that large shares of road transport will be electrified, including shares of the national truck fleet. In parallel with this, the Swedish transmission grid suffers from power capacity shortages, limiting the amount of electricity that can be distributed to a regional and local level, especially in urban areas, such as Stockholm. In line with this, the aim of this thesis is to investigate the trends of truck electrification in the Stockholm region and to assess its potential impact on the electrical grid based on truck operation characteristics. To achieve this, three objectives were set; to investigate truck fleet operators’ operations and view in relation to electrification, examine truck manufacturers' views on truck electrification and to analyse the truck electrification’s potential impact on the electrical grid in future scenarios. Quantitative and qualitative methods were used to fulfil the research objectives including interviews and casestudy modelling. The results from the interviews show that both manufacturers and truck fleet operators see the electrical grid and deployment of charging infrastructure, as well as the economy of the electric truck as the main barriers to overcome in relation to truck electrification. The truck manufacturers have taken on a leading role and are together with early adopters pushing the development forward. Nonetheless, they cannot create all the right conditions themselves, as for instance charging opportunities in relation to the electric grid is a complex problem. The industry agreed upon that city transports will be electrified first followed by regional transport and lastly long-distance transport. Furthermore, 2030 was identified as a key year for truck electrification. Finally, manufacturers and truck fleet operators urge the government to act and implement necessary measures to support the transition. The results from the case-study modelling show that lunchtime charging of city transport coincides with already critical hours for the electrical grid. Moreover, the afternoon charging of city, regional and long-distance transports generates a peak during the most critical hour around 6 PM, implying that it could entail challenges for an already congested grid. In line with this, proper night-time charging is considered as crucial both from an operational (in terms of minimising the daily stops) and grid point of view (in terms of avoiding grid congestion during critical hours). In addition, it is recommended to prioritise charging of long-distance transport during the day as they often are operative around-the-clock. Although this study resulted in general charging profiles, this gives a good indication on what impact the truck electrification might have on the electrical grid. Furthermore, it gives a general picture of how the electrification of trucks could play out in Stockholm, which can be applied in other urban areas in Sweden that are facing similar challenges. / I linje med Parisavtalet har Sverige satt upp nationella utsläpps reduktionsmål för transportsektorn; att reducera utsläppen med 70% till 2030, relativt nivåerna 2010. Detta medför att stora delar av transportsektorn kommer att elektrifieras, inklusive stora delar av den nationella lastbilsflottan. Parallellt med detta lider det svenska transmissionsnätet av kapacitetsbrist vilket begränsar mängden elektricitet som kan överföras till en regional och lokal nivå, speciellt i urbana områden såsom Stockholm. I linje med detta har syftet med denna studie varit att undersöka trender inom lastbils-elektrifiering i Stockholmsregionen och att utvärdera dess potentiella inverkan på elnätet baserat på lastbilars operativa mönster. För att uppnå detta sattes tre mål upp; att undersöka lastbils-operatörers verksamhets mönster och deras syn på elektrifiering, att undersöka lastbilstillverkares syn på elektrifiering och att analysera lastbils-elektrifieringens potentiella inverkan på energisystemet i framtida scenarier. Kvantitativa och kvalitativa metoder användes för att uppnå studiens mål inklusive; intervjuer och fallstudie modellering. Resultaten från intervjuerna visar att både lastbils-operatörer och tillverkare ser elnätet och utbredningen av laddinfrastruktur, såväl som totalekonomin av den elektriska lastbilen som de huvudsakliga barriärerna att överkomma vid en elektrifiering. Lastbilstillverkare har tagit en ledande roll och driver utvecklingen av elektriska lastbilar framåt med hjälp av så kallade early adopters. Däremot kan de inte skapa alla förutsättningar själva då till exempel laddnings möjligheter i relation till elnätet är en komplex fråga. Industrin är eniga om att city transporter kommer att elektrifieras först, följt av regionala och till sist fjärrtransporter. Vidare kunde 2030 identifieras som ett nyckelår för elektrifieringen. Slutligen uppmanar både tillverkare och lastbils operatörer regeringen att implementera nödvändiga åtgärder för att stötta och påskynda omställningen. Resultaten från modelleringen i fallstudien visar att lunch-laddning för city transporter sammanfaller med redan kritiska timmar för elnätet. Dessutom genererar eftermiddagsladdningen av city, regionala och fjärrtransporter, en effekttopp runt klockan 18 vilket skulle kunna medföra utmaningar för ett redan överbelastat elnät. I linje med detta anses natt-laddning vara avgörande både från en operativ synpunkt (i form av att minimera de dagliga stoppen) och från ett elnäts perspektiv (i form av att undvika överbelastning av nätet under redan kritiska timmar). Därtill, rekommenderas att fjärrtransporter prioriteras för dagtids-laddning då de ofta är operative dygnet runt. Trots att denna studie resulterade i generella ladd-profiler kan dessa ge en god indikation för vilken inverkan lastbils-elektrifieringen kan komma att ha på elnätet. Vidare, ger studien en generell bild över hur lastbils-elektrifieringen kan komma att se ut i Stockholmsregionen vilket kan tillämpas i andra urbana områden i Sverige som står inför liknande utmaningar.
9

Dynamic prediction of repair costs in heavy-duty trucks

Saigiridharan, Lakshidaa January 2020 (has links)
Pricing of repair and maintenance (R&M) contracts is one among the most important processes carried out at Scania. Predictions of repair costs at Scania are carried out using experience-based prediction methods which do not involve statistical methods for the computation of average repair costs for contracts terminated in the recent past. This method is difficult to apply for a reference population of rigid Scania trucks. Hence, the purpose of this study is to perform suitable statistical modelling to predict repair costs of four variants of rigid Scania trucks. The study gathers repair data from multiple sources and performs feature selection using the Akaike Information Criterion (AIC) to extract the most significant features that influence repair costs corresponding to each truck variant. The study proved to show that the inclusion of operational features as a factor could further influence the pricing of contracts. The hurdle Gamma model, which is widely used to handle zero inflations in Generalized Linear Models (GLMs), is used to train the data which consists of numerous zero and non-zero values. Due to the inherent hierarchical structure within the data expressed by individual chassis, a hierarchical hurdle Gamma model is also implemented. These two statistical models are found to perform much better than the experience-based prediction method. This evaluation is done using the mean absolute error (MAE) and root mean square error (RMSE) statistics. A final model comparison is conducted using the AIC to draw conclusions based on the goodness of fit and predictive performance of the two statistical models. On assessing the models using these statistics, the hierarchical hurdle Gamma model was found to perform predictions the best

Page generated in 0.0517 seconds