Transformer models, delivering big improvement in AI text-models (NLP), are now being applied in Knowledge Tracing to track the knowledge of students over time. One of the first, SAINT, showed quite some improvement over the then SOTA results on the public EdNet dataset and caused an increase in research based on transformer-based models. In this paper, we firstly aim to reproduce the SAINT results on the EdNet dataset but are unable to report a similar performance as the original paper. This might be due to implementation details, which we were not able to completely reconstruct. We hope to pave the road for further reproducibility, as an increasingly important part of AI research. Furthermore, we apply the model to a company dataset much larger than any public dataset (more interactions, more exercises and more skills). Such a dataset is on the one hand more challenging (more skills mixed), and on the other hand, provides much more data (which should help our models). We compare the SAINT model and the seminal IRT model, and find that the SAINT model performance is 4% better in AUC but 1.7% worse in RMSE. Our experiments on window size suggest that transformer models still struggle with modelling beyond recent performance, and do not yet deliver the step-change observed in NLP. / Transformermodeller, som ger stora förbättringar av AI-textmodeller (NLP), används nu i Knowledge Tracing för att spåra elevernas kunskaper över tid. En av de första, SAINT, visade en hel del förbättring jämfört med de dåvarande SOTA-resultaten på den offentliga EdNet-datauppsättningen och orsakade en ökning av forskning baserad på transformerbaserade modeller. I denna artikeln siktar vi först efter att återskapa SAINT-resultaten på EdNet-datauppsättningen, men vi kan inte rapportera liknande prestanda som den ursprungliga uppsatsen. Detta kan bero på implementeringsdetaljer som vi inte kunde rekonstruera helt. Vi hoppas kunna bana väg för ytterligare reproduktioner, som en allt viktigare del av AI-forskningen. Dessutom tillämpar vi modellen på en företagsdatauppsättning som är mycket större än någon offentlig datauppsättning (fler interaktioner, fler övningar och fler färdigheter). En sådan datauppsättning är å ena sidan mer utmanande (mer blandad kompetens), men å andra sidan ger den mycket mer data (vilket borde hjälpa våra modeller). Vi jämför SAINT-modellen och den framträdande IRT-modellen och finner att SAINT-modellens prestanda är 4% bättre i AUC men 1,7% sämre i RMSE. Våra experiment på fönsterstorlek tyder på att transformermodeller fortfarande kämpar med modellering utöver de senaste prestanda och ännu inte levererar den stegförändring som observerats i NLP.

Globalization has continuously brought an increased competition among companies, which entails a need for faster and more frequent deliveries of new products. Traditional project management methods, such as stage-gate and waterfall, are commonly used in production development projects and builds on a sequential approach. These methods have proven to have some disadvantages in flexibility, long lead times and it often creates communication barriers between the actors at each stage. The software industry has already encountered these obstacles and responded by introducing agile project management. Which improves the adaptability and allow changes to be made, due to new requirements from stakeholders or customers, throughout the entire development process. However, it remains unknown how agile models can improve production development. The purpose of this study was therefore to investigate how agile models can be applied to production development and what the effects are.  The authors have performed a case study at eight different companies within the automotive industry. The purpose of it has been to gain a deeper understanding about the case companies current production development processes and review how familiar the organizations are with the concept of agile project management. The extraction of the empirical data was conducted by questionnaires, interviews, and document reviews. An analyzation was done by comparing the empirical findings with the theoretical background out of eleven different categories that relates to project management (e.g., project goals, process, customer integration etc.). The analyzation concluded that the case company exclusively conducts their production development project by using a sequential approach.  The analyzation and the eleven categories where, together with the theoretical background about agile project management, later used to create the result by brainstorming different practices to become more agile. The results are presented out of three different scenarios, depending how agile the companies would like to be. For instance, are two process models suggested, one that is completely agile and one that is a hybrid of an agile and a stage-gate. Furthermore, are the implementation of self-organized teams, holistic approach towards internal and external partners, and reduced demand for documentation some of the practices that are suggested. Additionally, are three considerable aspects for the implementation presented.  The expected outcome and effects of applying these practices are discussed in the final chapter. Some of these outcomes are a company culture that will attract and retain talented personnel, where shared responsibilities and authorities triggers the employees to an increased commitment and sense of ownership towards their projects. Furthermore, are the companies expected to experience a more flexible and responsive approach towards conducting production development projects with a high focus on customer requirements and creating customer value.