Chalmers Open Digital Repository
Not a member yet
26247 research outputs found
Sort by
The Role of Maintenance Towards Circular Economy
No full textThe manufacturing industry is a major contributor to global greenhouse gas emissions, yet it also holds vast potential for sustainability transitions. As circular economy (CE) principles gain traction, integrating them into established industrial frameworks presents a persistent challenge especially within Company-Specific Production Systems (XPS) and its core pillars such as Professional Maintenance (PM) which are traditionally cost-driven and exclude social and environmental perspectives of sustainability. This thesis investigates how PM can evolve
into a strategic enabler of sustainability by embedding CE indicators.
Initially, a Systematic Literature Review (SLR) was conducted which identified 23 CE indicators at the micro and nano levels of manufacturing as well as common barriers and enablers to their deployment. These indicators were assessed for their measurability, data requirements, and strategic fit within PM through a structured questionnaire and semi-structured interviews with professionals across a Swedish manufacturing company.
The results showed that operational data was widely adopted at the sites, while energy, environmental and material data was significantly less adopted and more challenging for PM. Based on these insights, a parallel framework was developed that aligns the CE indicators with the existing PM maturity model. Together they capture both cost-efficiency and CE, by mapping indicators to maturity levels where data availability and implementation feasibility converge. Having created the framework, key deployment considerations were identified in the literature
and case study and then mapped against the appropriate levels in the framework. The study shows how management support and cultural readiness is important to consider initially, while more external perspectives, such as Environmental, Social, and Governance (ESG) reporting and supply chain collaborations are crucial towards higher maturity levels.
This thesis identifies CE Indicators suitable for PM, creates a CE-integrated maturity model
and identifies deployment considerations. By doing so it provides an academically based, yet
grounded in industry, way of integrating the CE perspective into PM
Atostugan - Cottage
No full textUnder tidigt 1920-tal byggde Tomas och Anna Klementsson en torvkåta nära Atoklimpen, i hjärtat av det samiska sommarvistet vid Risbäcken, Tärnaby. År 1925 färdigställde de ett trähus med kök, skafferi och rum på övervåningen—en banbrytande handling vid den tiden, då samer inte tilläts bygga permanenta bostäder i fjällområdet. Denna diskriminering ledde till att familjen kämpade för sin rätt att behålla huset, vilket bidrog till en lagändring år 1928 som gav samer rätt att uppföra fasta bostäder.
Intill huset upprättades en nomadskola för samiska barn, bestående av tre kåtor som användes för undervisning, måltider och övernattning. Den svenska regeringen såg dessa skolor som ett sätt att hålla samiska barn knutna till sin kultur och renskötseltradition. Skolan var i drift fram till 1945, då en ny skola byggdes i Tärnaby. Därefter blev familjen Klementssons hus bostad för läraren Anna Good
Starship Model 2025: Embedded System and Simulation for an Electrically Powered Reusable Rocket Model
No full textThe bachelor thesis describes the design, analysis, and simulation of an electrically powered reusable rocket model. The model is designed with two rotors mounted on tiltable arms, enabling lift and maneuverability. The simulation is performed in a close to realistic environment using MATLAB Simulink and Simscape. The primary focus is improving the modularity of the embedded system and simulating the model’s dynamic behavior.
The contributions made to the development of the Starship model are outlined at the start, detailing both hardware and software advancements. Key contributions include the design and implementation of a collision-resistant battery mount, a ground power cable for test bench applications, and the integration of a kill switch with multiple control modes. A planar motion detection system (optical flow sensor), a restructuring of embedded software, and position and orientation estimation using sensor data (lidar, IMU and Optical flow) are also developed.
The results demonstrate an accurate Complementary Filter for orientation estimation and a short term precise dead reckoning position estimation based on lidar, IMU, and Optical flow data. The optical flow camera provides the best position precision on rigid surfaces by identifying reference points. The restructuring of the software code results in a cleaner and more maintainable environment, enabling the seamless integration of additional features.
Additionally, the Simulink model accurately simulates the system’s behavior under applied forces. However, the results indicate that while short term stability is achieved, the design requires modification for long term stability due to physical limitations affecting its controllability
Kolmogorov-Arnold Nätverk, Framtidens AI?
No full textArtificiell intelligens integreras idag i alltfler tekniska system. Trots det, är det fortfarande
svårt att i praktiken implementera modeller från grunden på grund av höga resurskrav i form
av data och beräkningskraft. Detta motiverar sökandet efter effektivare modeller.
I denna artikel undersöks Kolmogorov Arnold Networks (KANs), en alternativ arkitektur
av neurala nätverk och jämförs med traditionella arkitekturer som multilayer perceptrons
(MLPs) och faltningsnätverk. KANs är intressanta då de potentiellt erbjuder fördelar som högre
noggrannhet, är lättare att tolka och innehar teoretiskt goda skalningslagar [1].
Vi undersöker KANs och särskilda modiferingar av dessa (Sprecher-KANs) relativt MLPs.
Jämförelsen görs genom att analysera utfallet från tre experiment. Specifikt analyseras mo dellernas förmågor för kurvanpassning, bildrekonstruktion och bildklassificering. I resultaten
uppvisar Sprecher-KANs högre prestanda än KAN och i sin tur att KAN överpresterar MLPs.
Faltningsnätverken överträffar KANs i prestanda, medan Sprecher-KANs presterar likvärdigt
med faltningsnätverken
Optimizing Bike Sharing Systems in Gothenburg: Enhancing Multimodal Public Transport Integration
No full textRentable bike sharing systems (BSS) are an environmentally friendly way of transportation
that has steadily gained popularity in cities worldwide in recent years.
These micromobility vehicles provide a convenient solution to the first-mile/lastmile
problem often associated with public transportation. While many studies have
focused on determining the optimal placement of BSS stations to fulfill user demand,
it is equally important to consider how well the BSS network integrates with
other modes of public transport to encourage multi-modal travel. The aim of this
thesis is to develop an optimization framework for determining the optimal placement
of BSS stations to enhance multi-modal accessibility while ensuring demand
fulfillment. The study focuses on the central parts of Gothenburg with demand
evaluated through analysis of historical travel data. The proposed framework defines
the problem as a facility location model, with the objective of maximizing a
defined accessibility measure. Constraints are introduced to ensure the fulfillment
of historical demand. Given the NP-hard nature of the problem, a metaheuristic
optimization algorithm is employed to find an efficient solution. The resulting station
placement is then compared to the current BSS station locations in Gothenburg
and demonstrates that the proposed solution improves the multi-modal accessibility
while successfully meeting historical demand
Measurement-Controlled Engines - Investigating the role of system-meter coupling time quantum information engines
No full textNanoscale devices that transform energy into useful work are becoming ubiquitous.
A critical task is to control energy transduction at the nanoscale. In this context,
quantum measurement and the associated information acquisition can be leveraged
to guide and enhance work output through feedback control. This thesis explores
a quantum information engine as a prototype energy-transducing device controlled
by measurement. This engine harnesses information transfer between a working
medium, modelled as a two-level system, and a meter, modelled as a quantum
harmonic oscillator. However, this information transfer is not instantaneous; it
depends on the coupling time, which is the time required to correlate the system
and the meter. This measurement time sets a lower bound on the cycle time of the
quantum information engine, making information acquisition a crucial resource for
the process.
We investigate the cost of quantum measurement, in particular the energetic cost
of coupling and decoupling the system and the meter in finite-time operations. Furthermore,
we analyse possible schemes of extracting useful work: the ergotropy, or
maximum work extraction under unitary transformations, and the excess work by
stimulated emission. In both cases, the information about the system is exploited
by conditioning the act of extracting work on the measurement outcome. Heat and
work flows are analysed as functions of the system and meter temperatures to show
that the quantum information engine can operate in different regimes: as a heat
engine, a heat valve, a refrigerator and a “true” information engine by extracting
work and cooling a colder bath. We show that the quantum information engine
performance in terms of power output for very short measurement times, the Zeno
limit, is small. To increase the power we need to increase the measurement time
which, however, comes with a higher cost of measurement. We carefully analyse the
work output-cost relation in different operating regimes of the quantum information
engine to find optimal conditions for net work output and high engine performance
Procedurally Generated Content in Games and its Effect on Player Experience
No full textThis thesis investigates the relationship between algorithmic complexity, procedural content complexity, and user experience for procedural content generation (PCG) in video games. It examines whether increased algorithmic sophistication leads to more varied and higher-quality content, and how such content compares to handcrafted levels in terms of player experience.
A prototype game was developed in Unity using C# to compare two PCG approaches, a basic random walk algorithm and a more advanced graph rewriting system. The generated game levels were evaluated using both player-centric and technical methods. Player experience was assessed through Think-Aloud play-testing, semi-structured interviews, the Game Experience Questionnaire (GEQ), and gameplay data analysis. Algorithmic complexity was determined using Halstead complexity measurements and Cyclomatic Complexity.
Results indicate that while the graph rewriting algorithm produced more diverse and engaging level structures than the random walk approach, it also introduced greater development complexity and reduced maintainability. Additionally, gameplay issues arose in procedurally generated content that were not present in handcrafted levels. These findings highlight the trade-offs between PCG complexity, content control, and content variety.
Ultimately, the thesis emphasises the need for controlled iteration, player-focused evaluation, and careful variable isolation in PCG system development. It provides actionable insights for game developers on when and how to apply procedural generation effectively while balancing content quality, development time, and system maintainability
Re-accessing: Recreating the corner tower as public building through methods of re-enactment
No full textThe thesis explores the corner tower as a prom
inent but inaccessible part of the urban fabric. The elaborate ornamentality that this typology
represents resembles the monument. But unlike the monument the corner tower is private, ac
cessible to the public only as a distant ornament. The private nature of the corner tower reduces
it to an anonymous object. Built as a symbol of status it is now a historical remnant without obvi
ous purpose, its visual identity unquestioned and spatial experience unexplored.
The project brings the corner tower to ground level, altered into an installation focusing both on
visual identity and spatial experience. The theory of re-enactment as explanation for architectural
development is applied in a series of methods to document and gradually alter the visual identity
of the corner tower. Through a new architectural work, the symbolic meaning as well as the possi
bilities of the corner tower is presented.
The aim of the thesis is to explore the methods related to architectural re-enactment by rein
terpreting the visual and spatial experience of the corner towers in a new work. The question
explored in the thesis is; How can values and meanings be transferred from an original to a new
work of architecture through methods of re-enactment?
A collection of corner towers in central Gothen burg is through the methods of quotation, para
phrase and spoliation documented, altered and repurposed in a new design project. Their spatial
possibilities are elaborated and their decorative elements are reinterpreted through experimental
design.
The design project consists of different installa tions, connected to the visual and symbolic val
ues of the corner towers. The corner tower has through the methods of re-enactment been rein
terpreted into a fortress, a series of follies and a watchtower. All three typologies are
contesting the inaccessible nature of the corner towers.
Translated to a new context, the designs rep resent a fragment of the architecture history of
Gothenburg. The locus is on the spatial possibili ties of the corner tower as well as its
ornamental nature. The result is a re-enactment of the cor ner tower, radically changed but still
connected to the same visual identity that inspired it
Effektivisering av containerflöde Värdeflödesanalys som verktyg för effektivare terminaldrift
No full textContainerterminaler utgör en central del av den globala logistikkedjan, där effek-
tivitet och tillförlitlighet är avgörande för konkurrenskraften. APM Terminals i
Göteborg, Skandinaviens största containerhamn, hanterar stora volymer last och
står inför utmaningar kopplade till hanteringen av tomcontainers. Dessa utmaningar
präglas av ojämn efterfrågan, obalanserad resursanvändning och ineffektiva flöden.
Detta examensarbete syftar till att kartlägga och analysera dagens containerflöde
i terminalen med hjälp av värdeflödeskartläggning för att identifiera icke-värdeska-
pande aktiviteter och föreslå åtgärder för att förbättra flödet. Genom en kombi-
nation av observationer, tidsstudier, dataanalys och intervjuer med nyckelpersoner
identifierades ineffektiviteter, framför allt väntetider, onödiga transporter och om-
arbeten.
Förbättringsförslag innefattar implementering av olika tidsbokningssystem, optime-
ring av terminalens layout samt ökad involvering av personal genom strukturerad
dialog. Dessutom föreslås specifika nyckeltal (KPI:er) som fångar upp återkommande
problemområden, exempelvis frekvensen av omarbetning och variation i efterfrågan.
Genom att rikta fokus från resurseffektivitet till flödeseffektivitet kan terminalens
verksamhet bli mer förutsägbar, flexibel och kundanpassad. Studien visar hur VFA
som metodik kan bidra till ett mer hållbart och konkurrenskraftigt arbetssätt i
komplexa logistiska miljöer
Dynamic Voltage Optimization for Energy Efficient Radios
No full textReducing energy waste in radios is essential for lowering the CO2 footprint of cellular connectivity. In current systems, radios operate at a static high voltage regardless of their physical resource block utilization (PRB-U), resulting in inefficiencies. This thesis proposes a machine learning (ML)-based approach to enable dynamic voltage control and evaluates which models and task formulations are most suitable for this purpose. Eight ML models were tested across two formulations: (1) classification of voltage levels and (2) regression to predict PRB-U values and map them to voltage levels.
The results show that ML can significantly reduce energy waste in certain radios, although effectiveness varies by device - some perform comparably using simpler methods. Classification outperformed regression in reducing voltage underestimations.
The Random Forest (RF) classifier and a customized Feedforward Neural Network (FFNN) classifier emerged as top performers. The FFNN achieved an F1 score of 0.35, saved 12.41% energy, and resulted in 34 underestimations. The RF reached an F1 score of 0.29, saved 12.43%, and had 42 underestimations. In contrast, the bestperforming baseline model had an F1 score of only 0.19 with 55 underestimations out of 68, underscoring the benefit of ML-based approaches.
Shifting the classification threshold helped manage the trade-off between energy savings and underestimations. Notably, the FFNN achieved just 5 underestimations while maintaining 10% energy savings.
SHAP-based explainability analysis showed that PRB-U at the current timestep was the most influential feature. The RF leveraged a broader feature set, while the FFNN focused more narrowly on a dominant input.
In conclusion, this thesis demonstrates that intelligent, ML-driven voltage control can enhance energy efficiency in radios. It also emphasizes the importance of balancing energy savings and prediction errors during model deployment, potentially by adjusting class probability thresholds