index.json

[{"content":"Francesco holds a PhD in experimental nuclear astrophysics from the University of Oslo, where he also continued as a postdoc before joining Expert Analytics. The subject of his PhD thesis was the extraction of nuclear properties from experimental data, and their use in astrophysical settings, such as the origin of heavy elements. He acquired expertise in nuclear physics and astrophysics, but also Python programming, statistics, experiment planning and data analysis. He likes solving complex problems, developing algorithms, and has keen interest in visualizing and presenting data.\nBefore his PhD, Francesco studied neutron stars at NTNU in Trondheim, Norway, and plasma physics at UiT in Tromsø, Norway. These are both complex subjects blending topics such as general relativity, thermodynamics, electrodynamics, fluid dynamics, quantum many-body systems and quantum field theory. During his academic journey, he developed skills in Python programming, data analysis, model simulations of complex nuclear and astrophysical systems, error propagation and data visualisation, as well as teaching, team collaboration and science communication.\n","permalink":"https://expertanalytics.no/about_us/employees/francesco/","summary":"","title":"Francesco Pogliano"},{"content":"Christian holds a PhD in fluid mechanics from the University of Oslo, where he also worked as a postdoctoral researcher. His expertise includes mathematical modeling, numerical methods and programming. Christian enjoys working in interdisciplinary environments and applying this knowledge to contribute to real-world solutions.\nDuring his PhD studies, he was involved in several research projects collaborating with multiple international research institutes. His main responsibility was the assessment of experimental data, including creating software for data analysis and visualization, the development of theoretical models, and the software implementation of numerical solvers. In his time in academia, he worked on problems related to elastohydrodynamics, lubrication, multiphase flow, charging of supercapacitors, and biomechanics. In addition, he has also worked with machine learning by combining neural networks with a\nreinforcement learning framework for flow optimization problems in tubular networks.\nBesides research, Christian is also an experienced lecturer and he has taught several courses in fluid mechanics at the University of Oslo as well as supervising students, both at the master\u0026rsquo;s and PhD levels.\n","permalink":"https://expertanalytics.no/about_us/employees/christian/","summary":"","title":"Christian Pedersen"},{"content":"Bastian holds a Ph.D. in computational mathematics from the University of Oslo. In his thesis he used mathematical modeling to investigate how molecules are transported in the human brain. His expertise lies in numerical simulations, nonlinear optimization and deep learning. Bastian is a systematic person with interest in computational modeling and working with imaging data.\nFrom his academic work, Bastian has gained substantial experience in combining large amounts of clinical imaging data with computational models. He is used to performing resource-demanding computations on supercomputers, including numerical simulations and training deep neural networks. Being a pragmatic programmer eager to learn new techniques, he used shape optimization methods in a data-driven approach to simplify the intricate workflow of his simulations.\n","permalink":"https://expertanalytics.no/about_us/employees/bastian/","summary":"","title":"Bastian Zapf"},{"content":"Ana holds a PhD in applied mathematics from University of Bergen and has worked as a postdoctoral researcher at Simula Research Lab in Norway. Her expertise lies in computational modeling, and algorithm design and implementation to solve large-scale problems in biomechanics and geophysics. Ana is an eager and challenge-driven person with strong interest in software development, scientific computing and machine learning.\nAna\u0026rsquo;s academic pursuits have primarily centered around numerical solvers for partial differential equations and the software implementation of computational techniques. Furthermore, she has delved into the analysis and utilization of neural networks, exploring their capacity to simulate physical laws and their performance in comparison to established methods.\nAna has also acquired hands-on experience in machine learning, statistics, and data visualization, particularly with social network analysis of sports data and image/video processing for motion detection.\n","permalink":"https://expertanalytics.no/about_us/employees/ana/","summary":"","title":"Ana Budiša"},{"content":"Eleonora obtained her PhD degree in applied mathematics from the University of Oslo and Simula Research Laboratory (Oslo, Norway) in 2023, and her Master\u0026rsquo;s degree in Aeronautical engineering from Politecnico di Milano in 2015. During her doctoral studies, she developed efficient solvers and robust formulations for poroelasticity equations applied to model brain displacements and fluid flow. From 2019 to 2023, Eleonora worked as a research and development engineer at FLIR Unmanned Aerial Systems AS (Hvalstad, Norway). In that role, she focused on improving the aerodynamical performances of drones using both numerical models and practical experiments.\nEleonora has experience in numerical simulations, mathematical modelling, testing, scientific writing, data acquisition, data analysis, and data visualization. In addition, Eleonora has supervised master students and summer interns providing guidance and feedback to help them produce high-quality work. Eleonora enjoys working in a collaborative and multidisciplinary enviroments. She is eager to continue learning new things while making valuable contributions in dynamic and demanding settings.\n","permalink":"https://expertanalytics.no/about_us/employees/eleonora/","summary":"","title":"Eleonora Piersanti"},{"content":"Kate holds a PhD in Marine Geology from Durham University, United Kingdom, and completed a post-doc at the University of Victoria, Canada. She is a problem solver, analytical thinker and loves a challenge. With her Earth Sciences background, she is an expert in big data analysis and visualisation, fieldwork planning and preparation, as well as instrument deployment. She has worked with a variety of geophysical data, and is always eager to dive into new data acquisition types. She has strong collaboration skills, extensive teaching experience and great communication skills.\nPrior to her PhD where she analysed Acoustic Doppler Current Profiler data, she completed her masters at Uppsala University, Sweden, with a specialisation in geomorphology and glaciology. Here, she processed and analysed Ground Penetrating Radar data from Svalbard, and did sediment analysis at an external laboratory in Denmark. She is an expert in the pipeline of field work design, data acquisition, data analysis through to data presentation.\n","permalink":"https://expertanalytics.no/about_us/employees/kate/","summary":"","title":"Kate Heerema"},{"content":"Eric holds a Ph.D. in Nuclear Physics from Berkeley. His thesis was written on the nuclear data of fission yields. His main interests include nuclear medicine, data science and analysis, and statistics. Eric is currently working with Expert Analytics to develop its oncology research portfolio after the company joined the Oslo Cancer Cluster in November 2022. Eric retains an appointment on the research faculty at the Department of Nuclear Engineering at Berkeley.\nEric completed his Ph.D. in August 2021 and thereafter joined the Department of Nuclear Engineering at Berkeley as a member of the research faculty. In October 2022, Eric joined Expert Analytics. He seeks projects that can utilize his skills in data science and nuclear science.\nEric\u0026rsquo;s main programming languages are Python and C++, however, he has extensive knowledge of FORTRAN and can also program in Java, MATLAB, and C. Eric has experience in open-source software development and open-access datasets. He is well-versed in using common scientific coding packages such as NumPy, SciPy, and ROOT.\nEric particularly enjoys developing thorough uncertainty quantifications for his analyses. These skills were developed in the course of his education and research in nuclear data. Eric frequently employs Monte Carlo methods for these quantifications.\nAs a result of his time as a National Nuclear Security Administration Fellow, Eric has developed an interest and considerable background in nuclear security policy. He led a student group at the Goldman School of Public Policy to publish their final course report in the Bulletin of Atomic Scientists.\nFinally, Eric greatly enjoys teaching. During his time at Berkeley, Eric was a teaching assistant or graduate student instructor on seven occasions. In 2018, he was awarded one of Berkeley\u0026rsquo;s \u0026ldquo;Outstanding Graduate Student Instructor\u0026rdquo; awards, a university-wide recognition.\n","permalink":"https://expertanalytics.no/about_us/employees/eric/","summary":"","title":"Eric Matthews"},{"content":"Therese holds a Ph.D. in Experimental Nuclear Physics from the University of Oslo. Her thesis was awarded the YARA Birkeland Prize for outstanding contributions to innovation and industry. Her main fields of expertise are developing real-time analytics tools and implementing machine learning and statistical models. The past years, she has developed prototypes and set into production software tools for the petroleum-, processing- and medical industry.\nDuring her Ph.D. work and following three-year postdoctoral fellowship, Therese planned experiments and analyzed results at Nuclear Physics labs in Japan and the US. Her programming languages of choice are Python and C++, but she also has experience writing code in MATLAB and Fortran.\n","permalink":"https://expertanalytics.no/about_us/employees/therese/","summary":"","title":"Therese Renstrøm"},{"content":"John is a computational scientist who has experience across disciplines in using machine learning, statistics, software development to reach solutions. John has worked on a variety of problems and enjoys learning new topics and new technology that connect across disciplines. At Expert Analytics John works on problems related to global scale spatio-temporal modeling and CO2 sequestration. John also is affiliated as Researcher at the University of Oslo in the Njord Centre for Studies of the Physics of the Earth.\n","permalink":"https://expertanalytics.no/about_us/employees/john/","summary":"","title":"John M. Aiken"},{"content":"Ricardo holds a PhD in applied mathematics from the University of Bergen and NORCE Norwegian Research Institute. With a strong foundation in theoretical and applied sciences, he has been a consultant since 2020, working in both public and private sectors. Ricardo\u0026rsquo;s skill set includes roles as a data analyst, data scientist, and developer. One of Ricardo\u0026rsquo;s key strengths is his expertise in Python and cloud computing, particularly in the Azure ecosystem.\nIn his academic pursuits, Ricardo delved into data assimilation algorithms, specializing in ensemble-based methods and data compression techniques mainly for oil and gas industry. Ricardo is passionate about applying this knowledge to contribute to real-world solutions.\n","permalink":"https://expertanalytics.no/about_us/employees/ricardo/","summary":"","title":"Ricardo Vasconcellos Soares"},{"content":"Felix is a research scientist with experience in data analysis and scientific programming. He holds a PhD in physics and worked in various interdisciplinary projects both as a Postdoc and Researcher in an academic setting as well as a senior researcher in industry. Felix has a background in both experimental studies as well as mathematical and numerical modelling. His creativity, analytical mindset and scientific experience provide him with a solid foundation for finding solutions to complex problems.\nHis areas of expertise include image processing, time series analysis, numerical simulations of complex physical systems, statistical data analysis, machine learning applications as well as visualisation and handling of large datasets. Felix worked in the fields of environmental physics, biophysics, optics, microscopy, physics of complex systems, crystal growth, geophysics, neuroscience, sport science and medical instrument technology. During his career he collaborated with people from various backgrounds including meteorology, medicine, biology, engineering, geology, psychology and biochemistry. Felix is driven by curiosity and is always looking for new challenges.\n","permalink":"https://expertanalytics.no/about_us/employees/felix/","summary":"","title":"Felix Kohler"},{"content":"Alessandro has a PhD in Physics and 10+ years of corporate and academic experience in software development, machine learning, scientific research, and technical troubleshooting. Alessandro is eager to work with new technologies and is experienced in solving problems analytically and with creativity. At the same time he is proficient in handling technical issues and in communicating with end users.\n","permalink":"https://expertanalytics.no/about_us/employees/alessandro/","summary":"","title":"Alessandro Marin"},{"content":"Kine is an organized scientific programmer that likes to solve problems from a big picture approach. She submitted her PhD thesis in Climate Science in December of 2021, and during her doctorate she analyzed radiation and air pollution data from a multitude of climate model simulations and compared them with observations. This work gave insight in big data analytics and visualisation, especially regarding time series and how to avoid common pitfalls like comparing apples and oranges. She enjoys working with large datasets, learning new things, and communicating/teaching science.\nKine gained HPC and scripting experience from designing and performing climate simulations with the Norwegian Earth System Model during her PhD, and wrote the current introductory documentation (newbie guide) for this model. She has strong communicative skills and enjoys experimenting with data visualization as a communicative tool.\nShe values collaboration and teamwork, and she has experience in working with diverse teams through academic research projects. Her research paper on climate model performance (Moseid et al., 2020) has been recognized by being cited in the United Nations Intergovernmental Panel on Climate Change (IPCC) newest physical assessment report.\nIn her spare time she enjoys taking pictures of clouds.\n","permalink":"https://expertanalytics.no/about_us/employees/kine/","summary":"","title":"Kine Onsum Moseid"},{"content":"Kristine has a Master\u0026rsquo;s degree and three years experience as a PhD-student in Nuclear Physics. She has accumulated extensive practice with statistical analysis of sensor data and problem solving using programming tools such as Python and C++. Kristine has performed analyses consisting of handling raw data, performing calibrations, doing time series analysis, statistical fitting of peak distributions and doing model simulations in order to compare to experimental data.\nWith a background in statistics, machine learning and Bayesian statistics she is able to extract meaningful information and interpret characteristics from data. She always approaches challenges with eagerness and can efficiently learn new skills in order to solve unique problems.\n","permalink":"https://expertanalytics.no/about_us/employees/kristine/","summary":"","title":"Kristine Sønstevold Beckmann"},{"content":"Marco holds a Ph.D in computational Condensed Matter Physics, with a strong focus on quantum and classical atomistic simulation methods. Throughout his decade-long career he has worked on both software development - mainly in C++ and Python - and research intensive projects across several research fields ranging from low-temperature physics to computational chemistry and biophysics. This experience, along with his robust mathematical and coding background, allows him to be a broad and thorough problem solver, well capable of working both independently and as a team member.\nSince 2005 Marco worked on a rich variety of projects. To name a few, he worked on a porting of a Quantum Monte Carlo simulation software from Fortan77 to C++. He later designed and developed parallel simulation software for quantum many-body systems adsorbed on various substrates, tools for the solution of inverse problems (i.e. the inversion of the numerical Laplace transform from noisy data) based on Genetic Algorithms, atomistic simulation libraries for chemical systems with hybrid openMP/MPICH parallelization, non-linear fitting strategies for Molecular Dynamics thermostats based on generalizations of the Langevin Equation, a Python script that implements some Enhanced Sampling methods in Gromacs, and more recently coordinated the development of a C++14 program for quantum dynamical simulations based on the Feynman\u0026rsquo;s Path Integral. He worked on several physical and mathematical problems including Superfluidity and Supersolidity, the negative sign problem affecting sampling algorithms, Enhanced Sampling of high dimensional probability distributions, and the problem of quantum transport in protein channels.\n","permalink":"https://expertanalytics.no/about_us/employees/marco/","summary":"","title":"Marco Nava"},{"content":"Mats has a PhD (2019) in applied and computational mathematics (University of Bergen, Bergen, Norway), in addition to a postdoctoral experience in biomathematics at the University of Oslo, Oslo, Norway (2020-2021). He has experience and interest in biomathematics, fluid mechanics, flow in porous media, heat transfer, solid mechanics, coupled systems, among others. Moreover, he has programming experience in Python, Java, Matlab and Haskell. Mats is an analytical and creative problem solver, whose expertise relies on diverse experience with mathematical modeling and numerical programming, with applications ranging from geothermal energy storage to evolutionary biology, in addition to several teaching experiences.\n","permalink":"https://expertanalytics.no/about_us/employees/mats/","summary":"","title":"Mats Brun"},{"content":"Sebastian is a curious person always looking to learn about new topics and combining them into novel solutions to complex problems. He received his PhD in biocomputional sciences in March 2021 from the Italian Institute of Technology/University of Bologna, where he studied how nanomaterials interact with physiological environments. He is experienced in molecular simulations, data visualization, and scientific writing. In more recent years, Sebastian has been applying bioinformatic methods to genomics, proteomics, and transcriptomics data.\nDuring his PhD, Sebastian managed large amounts of data (hundreds of gigabytes) generated from molecular simulations. In this type of computation, numerical methods are used to iteratively solve coupled differential equations, which in turn offers an atomic description of physicochemical processes.\nWhen approaching a scientific problem, Sebastian looks for inspiration in Nature, which has perfected behaviors (e.g. preying strategies) over millions of years.\nSebastian is particularly drawn to creative, unconventional ways of displaying data. He is passionate about generative art, dancing, and animals. His values are based on critical thinking, diversity, inclusion, and teaching.\n","permalink":"https://expertanalytics.no/about_us/employees/sebas/","summary":"","title":"Sebastian Franco Ulloa"},{"content":"Thomas submitted his PhD thesis in Applied Mathematics in November 2020 to the University of Oslo, with supervision from NGI Oslo. His thesis deals with the effect of submarine landslide failure and flow on tsunami genesis. This topic gave him valuable insights into the complexity of this natural disaster. Through numerical modelling, he developed strong scripting skills in Python and Bash.\nHis Master education at ETH Zurich, with Major in Geophysics, let Thomas understand processes and structures of the Earth\u0026rsquo;s interior. He was educated to obtain a solid understanding of mathematical, numerical, and physical methods within a range of geophysical disciplines such as seismology, geodynamics, and applied geophysics.\nDuring the Bachelor education at ETH Zurich and the University of Toulouse he learned about the major concepts and methodologies of Earth Sciences, and developed independent work and problem-solving skills.\nThomas is an analytical thinker and ready to deliver digital solutions for the clients.\n","permalink":"https://expertanalytics.no/about_us/employees/tzm/","summary":"","title":"Thomas Zengaffinen-Morris"},{"content":"Max has a PhD (2014) in Aerospace Engineering (INPE, Brazil) and two postdoctoral experiences: one in Applied Mathematics, at IMPA (Brazil, 2014-2016) and another in Chemistry, at the University of Bergen (2016-2019). He has experience and interest in multiphase flow, reactive flows, flows in porous media, heat transfer, CFD, to name a few. Additionally, he has programming experience in Fortran, Matlab, Python and Octave.\nMore than the scientific capabilities, Max\u0026rsquo;s expertise relies on its experience in bridging fundamental and applied science.\nDuring the five years Max was a postdoctoral fellow, he collaborated with researchers from several countries (Brazil, USA, Netherlands, Norway, Belgium, Sweden, UK) and different backgrounds (mathematicias, phyisicists, chemists, engineers).\n","permalink":"https://expertanalytics.no/about_us/employees/max/","summary":"","title":"Max Endo Kokubun"},{"content":"Sigmund submitted his masters\u0026rsquo;s degree in Geophysics and Seismology at the University of Oslo in 2019. The subject of the thesis was Attenuation of Seismic Interference Noise using Convolutional Neural Networks and was written in collaboration with CGG Services AS. This work gave insight in the fields of machine learning and seismic signal processing.\nWorking with industrial problems utilizing machine learning has given him experience with convolutional neural networks and seismic data processing\nand further led to a consultancy job at Lundin Norway AS working on related projects. He has mainly used Python and has experience with machine learning frameworks such as TensorFlow and Keras, as well as numerical libraries such a Numpy. He is also familiar with MATLAB, C++ and Bash.\n","permalink":"https://expertanalytics.no/about_us/employees/sigmund/","summary":"","title":"Sigmund Slang"},{"content":"Robert Solli is a specialist in scientific programming, with expertise in mathematical optimisation, statistical analysis and machine learning. He has solved a varied set of problems with this skillset, from understanding complex physical systems to increasing student volunteer participation.\nRobert submitted his master\u0026rsquo;s degree in Computational Physics in 2019. In the thesis, Robert analysed events in a nuclear physics experiment to separate reaction products using machine learning techniques. He completed the degree at the University of Oslo where he also worked as a research assistant. In his work as a research assistant, he analysed and wrote a paper on understanding student interactions with an online video course.\nRobert is also an avid climber and takes as much delight in finding solutions to hard problems on a sheer rock-face as to those presented by data.\n","permalink":"https://expertanalytics.no/about_us/employees/robert/","summary":"","title":"Robert Solli"},{"content":"Robert has a PhD in astrophysics, with a specialisation in cosmological simulations. He is experienced with numerical modelling of physical systems, statistical analysis of large data sets, high-performance computing, and cloud based applications.\nRobert finished his PhD at the University of Oslo in 2019. His thesis involved solving highly non-linear systems of differential equations with parallel computing and analysing gigabytes of output data. His scientific background and curious mind is a strong combination for finding and synthesizing knowledge, as well as for quickly adapting to new situations. In addition to science and programming, he has a passion for teaching and explaining technical concepts in an easily understandable way.\n","permalink":"https://expertanalytics.no/about_us/employees/rh/","summary":"","title":"Robert Hagala"},{"content":"Ada holds a PhD in Physics (specialization in Astrophysics) from the University of Barcelona, Spain, which she obtained in 2003. She has 20 years of experience in academia doing research in the field of Solar Physics. This background provided her with experience in handling large datasets from different spacecrafts and ground-based observatories, scientific programming, image analysis, time-series analysis, analytical modelling, signal processing, as well as the development of feature recognition algorithms and visualization tools for large volumes of data. Such skills can also be applied to complex problems in the private industry.\nAda has an analytical mind, which together with thoroughness and perseverance provides her with the ability to solve problems. Her interests expand beyond the purely scientific domain into education, scientific communication and outreach, and organization. Her peers regard her as an excellent communicator and an all-terrain asset.\n","permalink":"https://expertanalytics.no/about_us/employees/ada/","summary":"","title":"Ada Ortiz-Carbonell"},{"content":"Ata holds a PhD in physics with specialisation in astrophysics which he obtained from Brown University in 2014. Before joining the Expert Analytics team, he has worked as a postdoctoral researcher at the University of Oslo and Universite Paris VII. His expertise includes statistical analysis of large data sets, numerical modelling, signal processing, imaging, and programming.\nHis strong analytical skills and scientific background together with a demonstrated ability to learn complex material provide him to apply his expertise on finding solutions for multifaceted complex problems.\n","permalink":"https://expertanalytics.no/about_us/employees/ata/","summary":"","title":"Ata Karakci"},{"content":"Thomas is interested in bringing to life the potential of data for automation and insight purposes. He will happily engage in the various steps of the process, from analysis and machine learning to data collection and software development.\nThomas holds a PhD in astrophysics for developing and studying models of the solar atmosphere. After finishing the PhD he worked as an analyst in the public sector. Here he took part in the business intelligence process as well as data analyses, data modeling and system architecting. Thomas also completed the two year leadership development program Teach First Norway.\nThe varied experience makes Thomas a diverse problem solver with a critical and systematic approach. He has the ability of quickly adopting to new projects and suggest novel solutions.\n","permalink":"https://expertanalytics.no/about_us/employees/thomas/","summary":"","title":"Thomas Golding"},{"content":"Diako holds a Ph.D. in Computational Mathematics from the University of Oslo. As a student at both Mathematics and Physics departments at the University of Oslo, he has acquired a broad knowledge in various physical and mathematical theories, and numerical methods.\nDuring his Ph.D. period he has been involved in several projects within multiple research groups at the Faculty of Mathematics and Natural Sciences. He has developed high performance numerical codes, in C/C++ and Python/Cython combining various numerical methods, to study plasma-object interactions and ionospheric instabilities and turbulence. In addition, he has developed new analytical models describing the interactions between electrically conducting objects and plasmas.\nThrough his education, Diako has developed strong analytical and numerical skills to solve complex problems related to phenomena arising in the natural world.\n","permalink":"https://expertanalytics.no/about_us/employees/diako/","summary":"","title":"Diako Darian"},{"content":"Simen is a computational scientist who finished his Ph.D. in computational neuroscience in 2019. He has a background from computational astrophysics where he wrote software for finding clusters of galaxies in cosmological N-body simulations. His Ph.D. focused on quantifying uncertainties in computational models of neurons and neural networks and he created a Python toolbox for performing these calculations.\nAdditionally, the work included creating a new specification for data storage, data analysis, and developing an educational neural network simulator. Simen also has teaching experience and has co-authored a textbook that teaches first-semester bachelor students in biology programming, data analysis, and computational modeling.\nSimen has experience at learning new disciplines and enjoys using the computer to solve complex problems, create models and numerical simulations and perform data analysis.\n","permalink":"https://expertanalytics.no/about_us/employees/simen/","summary":"","title":"Simen Tennøe"},{"content":"Kent-Andre is a Professor of Applied Mathematics at the University of Oslo. He has extensive experience with computational modeling, multi-physics simulations, high-performance computing, and scientific software development (Python and C++).\nKent-Andre is experienced in multidisciplinary collaborations where advanced computational techniques are developed to solve real-life problems, in particular in the medicine. In XAL, he brings a broad experience in inter-disciplinary work as well as technical experience in scientific computing involving big data.\n","permalink":"https://expertanalytics.no/about_us/employees/kent/","summary":"","title":"Kent-Andre Mardal"},{"content":"Alocias has a master\u0026rsquo;s degree in physics from the University of Oslo, submitted in spring 2018. The subject of the thesis was Quantum Monte Carlo Simulations of Quantum dots constrained in single- and double well potentials. He solved the numerical problem for both systems with new analytic expressions which had not been explicitly done before and a newer method with roots in machine learning to further improve upon the results.\nSolving these complex systems gave him a varied skill set within the fields of numerical methods, statistical analysis, mathematical optimization, symbolic computation and high performance computing. This has made him experienced in programming with multiple languages and skilled in efficient algorithmic approaches.\nWith his education in Computational Physics and a curious mindset, Alocias is a positive learner and well-adapted for solving complex problems.\n","permalink":"https://expertanalytics.no/about_us/employees/alocias/","summary":"","title":"Alocias Mariadason"},{"content":"Roar has a master\u0026rsquo;s degree in High Energy Physics from the University of Oslo, completed in 2018. In his thesis he studied two different models of particle production at high energies, one statistical using the thermodynamical equations and one using the properties of a theoretical particle called Pomeron.\nHe also has a practical background working as an electrician. All through his studies he worked with C, C++ and Python, but is adaptable to new systems and languages.\n","permalink":"https://expertanalytics.no/about_us/employees/roar/","summary":"","title":"Roar Emaus"},{"content":"Sebastian holds a master\u0026rsquo;s degree in applied mathematics from the University of Oslo and Simula Research Laboratory, finished in 2017. In his thesis he built a Fluid-Structure Interaction solver with Python using the FEniCS platform. This also included working with supercomputers parallelizing the computational efforts. During his studies he gained knowledge in algorithms, numerical computing and programming in different languages.\nAfter graduating he worked for Shortcut AS, an app-development company, developing iOS apps and working with Splunk for monitoring and analyzing machine-generated big data. He has also taken courses and attended conferences on AI/Machine Learning, learning neural networks using Tensorflow, Keras, scikit-learn.\nApart from that he is highly adaptable to any situation and can quickly learn new languages and frameworks.\n","permalink":"https://expertanalytics.no/about_us/employees/sebastian/","summary":"","title":"Sebastian Gjertsen"},{"content":"Alexander is a senior software developer specializing in backend and system development in Python. His experience with Python dates back to 2010, and he has been working professionally with system development in Python since 2018. During his career, he has worked in several platform and product teams in both the public transport sector and the energy sector. Alexander holds a master\u0026rsquo;s thesis in computational quantum physics from the University of Oslo.\nThe subject of Alexander\u0026rsquo;s thesis was Monte Carlo simulations of quantum dots using C++ and Python.\nHis education has trained him in solving complex physical and mathematical problems using various numerical methods and libraries as well as visualization. He is also a fast learner and can adapt to new frameworks and programming languages quickly.\n","permalink":"https://expertanalytics.no/about_us/employees/alexander/","summary":"","title":"Alexander Fleischer"},{"content":"Guttorm is a curious person with an interest and enthusiasm for technology and problem solving. His academic background is focused around applied mathematics, programming, and physics. He submitted his master’s thesis the summer 2017 in the field of Computational Science and Engineering at the University of Oslo / Simula.\nIn his work he performed numerical simulations of micro-particles inhaled into the human respiratory system. This involved writing Python scripts as well as modifying and working with open-source libraries, in addition to data analysis and supercomputer simulations.\nWhile working at Expert Analytics, he has continued developing his programming skills in Python and Go, and has been diving into topics such as machine learning and web programming.\n","permalink":"https://expertanalytics.no/about_us/employees/guttorm/","summary":"","title":"Guttorm Kvaal"},{"content":"Jakob holds a Ph.D. in computational mathematics from the University of Oslo, where he studied mathematical models of electrical activity in the brain. He also developed computer models running on high-performance computers produces large amounts of data. Jakob is a pragmatic programmer who takes pride in building robust, clean and easy-to-maintain software solutions. Since being awarded his Ph.D, Jakob has built and maintained energy-market models in the renewable power sector.\nJakob\u0026rsquo;s experience working in an interdisciplinary field, and enthusiasm for learning have been key to his success in collaberating with experts from diverse fields. By working closely with colleagues, Jakob has been able to deliver practical and technically sound solutions that meet real-world needs. He values the importance of learning from others and sharing his own expertise to support others, and believes that the best solutions are achieved in collaboration with peers with a range skills and experiences.\n","permalink":"https://expertanalytics.no/about_us/employees/jakob/","summary":"","title":"Jakob Schreiner"},{"content":"Eivind has a masters degree in Computational Science from the University of Oslo, completed in 2016. His project investigated a particle system modeling linear elasticity, and accelerating linear algebra computations using GPUs. From the studies leading to his degree he has gained broad knowledge about algorithms, numerical mathematics, and programming in several languages.\nPrior work experience includes teaching programming: mostly one-to-one teaching in a lab setting, but also as lecturing and live coding in plenary settings. He has also worked with testing\u0026mdash;both manual and automated, and with Android development and some minor Javascript development.\n","permalink":"https://expertanalytics.no/about_us/employees/eivind/","summary":"","title":"Eivind Storm Aarnæs"},{"content":"Vinzenz is a curious, pragmatic, creative problem solver. He holds a PhD in biomechanical engineering, focusing on stochastic simulations of the blood flow in the human arterial system. This interdisciplinary work included working within engineering, medicine, biology, software development and statistics.\nHis educational background is in mechanical engineering, with focus on numerical methods and bio-materials. Beside his studies, he has worked in several different fields such as virtual commissioning, petroleum research, biomechanics research. His educational and work experience gives him strong analytic skills, the ability to see problems from different angles and has strengthened his communications skills.\nIn the last years he has mainly been programming in Python, which he has used to write a comprehensive simulation software. He is also familiar with other languages, like C and C++, and can easily adapt beyond this.\n","permalink":"https://expertanalytics.no/about_us/employees/vinzenz/","summary":"","title":"Vinzenz Gregor Eck"},{"content":"Ola is the CEO of Expert Analytics. He has a PhD in scientific computing, and is an expert software developer with decades of experience in industry and academia. He is particularly interested in mathematical and numerical software development, algorithm development, high performance computing and scientific visualization.\n","permalink":"https://expertanalytics.no/about_us/employees/ola/","summary":"","title":"Ola Skavhaug"},{"content":"4DModeller is a spatio-temporal modelling package that can be applied to problems at any scale from micro to processes that operate at a global scale. It includes data visualization tools, finite element mesh building tools, Bayesian hierarchical modelling based on Bayesian inference packages INLA and inlabru, and model evaluation and assessment tools.\n4DModeller has been designed to make it easy to design spatially distributed, temporally dependent statistical models. Typically, 4DModeller expects tabular data sets with spatial coordinates, time indices, and the values that change or remain constant over those times. It is designed to be used in the modelling process once data has been sufficiently organized from wherever it was gathered from.\nThis package is typically used to model continuous processes (e.g., sea level rises, earth\u0026rsquo;s magnetic field), probability distributions of point processes (e.g., earthquake locations), or residuals.\n4D Modeller was developed in partnership with University of Bristol, University of Oslo Njord Center, TU Munich and Expert Analytics.\nCheck out the dedicated page for 4D modeller!\nhttps://4dmodeller.github.io/fdmr/\n","permalink":"https://expertanalytics.no/rd/4d/","summary":"Seamlessly model continuous processes in R!","title":"4D Modeller"},{"content":"Long term power production planning involves resolving a comprehensive market model with power consumers, producers, and transmission lines, down to the individual reservoirs and hydro power stations and pumps in coupled water courses. This is done by running simulation software that outputs a fairly large amount of data in the range of 15 - 20 GB each time a simulation is run.\nWe have helped building an analysis platform for this problem that contains a domain model in which simulation results are inserted after they are computed. Furthermore, we have developed advanced techniques for retrieving and visualizing the results in terms of time series data and statistics, along with advanced aggregation of results to allow forecasts to be viewed in ways earlier not feasible.\n","permalink":"https://expertanalytics.no/use_cases/adam/","summary":"Domain modelling of the hydro thermic electrical system in order to allow advanced context based analysis and visualization.","title":"Adam"},{"content":"For one of our FinTech clients, we were asked to improve their rule-based transaction monitoring system to reduce the false positives in the detected transactions.This would ultimately reduce the number of cases the case workers had to go through each day.\nWe leveraged an AI driven technique to classify output of the transaction monitoring system and report a probability score on the conviction of each case. Additionally, we implemented reasoning techniques to elaborate on the decision made by the model for transparency and in order to be able to continuously refine and enhance our model performance. A full MLOps setup was implemented alongside the models with daily training, model evaluation and serving the model to production.\nIn the first phase of the deployment, we set a conservative threshold for automatically closing the cases which are considered highly unlikely to be convicted by the model. Already this low threshold removed cases equal to the performance of 1 case worker in a week.\n","permalink":"https://expertanalytics.no/use_cases/ai_fraud/","summary":"Using AI to lower the false-positives in detected fraudulent transactions","title":"AI-based Financial Fraud Evaluation Assistant"},{"content":"This project focused on developing a comprehensive data platform designed to support the collection, analysis, and modeling of large datasets for energy management. The platform integrates legacy systems with modern cloud solutions, enabling seamless data flows for forecasting and predictive maintenance. It handles third-party data sources, such as meteorological data, and internal measurement data, providing the foundation for advanced analytics that drive critical business decisions and operational improvements.\nOur team played a key role in developing predictive maintenance solutions that significantly extended the lifespan of vital equipment. By leveraging advanced statistical models and machine learning algorithms, the platform could identify potential equipment failures before they occur, allowing for timely interventions. This proactive maintenance approach resulted in reduced downtime and optimized the performance of essential energy assets, ultimately enhancing operational efficiency and extending equipment longevity.\nWe implemented an end-to-end monitoring system powered by AI, integrated with the client’s infrastructure, and designed to provide real-time insights. Utilizing a combination of technologies, such as Python, Azure, and containerized microservices, our team ensured the platform’s scalability and reliability. The solution helped improve decision-making in energy management and led to significant cost savings by preventing equipment failures and allowing for higher performance under optimized conditions.\n","permalink":"https://expertanalytics.no/use_cases/predictive_maintenance/","summary":"Predicting wear, extending life: intelligent maintenance for lasting performance","title":"AI-Powered Predictive Maintenance"},{"content":"If you\u0026rsquo;d like to get in touch with us, please email us at post@expertanalytics.no or call us at +47 926 12 490.\nIf you are the Oslo area, feel free to visit our office in Mesh Workspace, located at Møllergata 8\n","permalink":"https://expertanalytics.no/contact/","summary":"How can we help you?","title":"Contact Us"},{"content":"Expert Analytics is a member of the Oslo Cancer Cluster. Our goal as a member is to leverage our software development and data analytics skills to help the health sector in devising strategies for cancer treatment.\nAs members of the OCC, we were approached by the CellFit Project, a consortium of collaboration between Oslo University Hospital, Oslo Cancer Cluster, SINTEF and Thermo Fisher Scientific, which addresses cell-based cancer therapy. From this approach, the CytoFit project was born!\nCytoFit is a software solution that automates the processing of cytometry data. Despite being a key component of personalized precision medicine, cytometry data analysis often involves subjective user input that compromises the robustness of the results. Cytofit aims to streamline the storage, preprocessing, analysis, and visualization of these data in a user-friendly cloud-based application, enhancing collaboration between clinical laboratories, healthcare professionals, and researchers. The software addresses inefficiencies in current systems by integrating advanced analysis models and high-quality visualizations, ultimately improving the quality of healthcare and accelerating the development of patient-centered therapies.\nWe are using Cytofit to analyze proteomics datasets from Oslo University Hospital and optimize experimental cancer treatments. This is accomplished by profiling the immune system of individual patients and look for biomarkers that predict clinical outcomes like treatment responsiveness.\nWe are thrilled to utilise our scientific and software background to develop meaningful solution in the fight against cancer!\nThe project was presented in The Intelligent Health Conference 2024 (\u0026ldquo;The CellFit Project\u0026rdquo;).\n","permalink":"https://expertanalytics.no/rd/occ/","summary":"Cytofit is a Python library to process Omics data and draw meaningful insights for the development of personalized medical treatments.","title":"CytoFit"},{"content":"For Statnett, we have assisted in two strategically important projects. First, by making compiled data accessible to Statnett\u0026rsquo;s team of analysts in the planning and operations environments. Later, as part of the team during the implementation of flow-based market coupling.\nIn both projects, we contributed Data Science resources. For flow-based market coupling, we leveraged strong scientific expertise to contribute to the design of quality improvement efforts for the project. In the development of the platform and service for making data accessible, we also contributed with system development expertise and technical leadership. To make data available in the simplest possible way required intimate knowledge of complex industry standards, which had to be simplified to enable an intuitive approach for an analyst.\nBoth projects were highly prioritized and strategically important for Statnett, and our collaboration has contributed to the high professional quality of the deliveries we supported, while also ensuring highly efficient use of resources.\n","permalink":"https://expertanalytics.no/use_cases/statnett/","summary":"Enhancing efficiency in the energy market through Data Science and System Innovation","title":"Data Enablement \u0026 Flow-Based Market Integration"},{"content":"Data Engineering is the process of designing, constructing, and maintaining the architecture that enables the collection, storage, and analysis of large volumes of data. It involves the development of data pipelines that automate the extraction, transformation, and loading (ETL) of data from various sources into a central repository. This ensures that data is accurate, consistent, and readily available for analysis and decision-making. Data engineers use a variety of tools and technologies to build scalable and efficient systems, allowing businesses to manage their data assets effectively and support their analytics and machine learning initiatives.\nFor businesses aiming to maximize their operations, data engineering is crucial. It ensures that high-quality data is available for analysis, enabling more accurate and timely insights. By automating data workflows, businesses can reduce manual errors, save time, and increase productivity. This leads to better decision-making, improved customer experiences, and enhanced operational efficiency. In a data-driven world, robust data engineering practices empower businesses to leverage their data assets fully, stay competitive, and drive innovation.\n","permalink":"https://expertanalytics.no/services/data_engineering/","summary":"Elevate your data operations with cutting-edge automation, ensuring efficient, consistent, and reliable data processing. Optimize performance and scalability while minimizing manual errors, keeping your data operations agile and robust.","title":"Data Engineering"},{"content":"Data Science is the science (and art) of extracting knowledge or insight from raw data. Extracting meaning from and interpreting data requires tools and methods of multiple disciplines such as mathematics, statistics, physics and computer science.\nIn Expert Analytics we apply scientific methods as a pillar to obtain valuable knowledge out of data. In combination with industry-grade programming methodology, our consultants craft solutions which are tailormade for your business needs!\nExtracting value from data can benefit your organisation in multiple ways: from forecasting unforeseen events and detect failures before they occur, to automate process and obtain significant and measurable KPIs for your business!\n","permalink":"https://expertanalytics.no/services/data_science/","summary":"Harness the power of data science and machine learning with our team of highly educated experts. We transform raw data into actionable insights, driving innovation and informed decision-making. Let us help you unlock the potential of your data to achieve your business goals!","title":"Data Science \u0026 Business Intelligence"},{"content":"The health of your equipment can be measured by the sound it makes. This fact, exploited by technicians for decades, inspired our journey to develop our edge audio analytics system that will record and learn from the machine sounds. The result is a robust and performant monitoring system that will alert you when sudden events or creeping changes require your attention.\nOur Edge Audio Analytics solution is capable of continuously recording and analysing up to 16 channels of 48kHz each on the edge hardware, and is equipped with a cloud agnostic control system utilising MQTT as the communication protocol.\nOn the edge, we run tens of models to capture the audio features and understand them. These models can be simple, like calculating the total loudness from a microphone, or advanced machine learning models used to classify certain audio patterns of interest. Only the relevant information, in terms of time series and audio snippets are uploaded to the cloud, reducing the cloud cost by several orders of magnitude.\nOur Edge Audio Analytics solution was developed in partnership with Å Energi, StepSolutions, with partial funding from Innovasjon Norge.\nExpert Analytics was part of the Distributed Artificial Intelligence Systems (DAIS) initiative, which is a pan-European project aimed at bringing faster and more secure data processing solutions through the development of edge AI software and hardware components. This large consortium was funded by the European Union\u0026rsquo;s Horizon 2020 research and innovation program.\nCheck our website dedicated for the Edge Audio Analytics solution!\nhttps://www.audioanalytics.de/\n","permalink":"https://expertanalytics.no/rd/audio/","summary":"Non-invasive monitoring for your assets powered by state-of-the-art AI","title":"Edge Audio Analytics"},{"content":"Norgips, a Norwegian producer and supplier of gypsum boards across Norway and Europe, aimed to enhance its production processes by utilizing sensor data collected from its factory. The primary challenge was to transform this vast stream of data into actionable insights to reduce costs and improve product quality.\nApproach To meet this challenge, we designed and implemented an analytics pipeline that consolidated and processed sensor data into actionable insights. This involved orchestrating machine learning models in both production and development environments to analyze patterns and predict quality metrics. Additionally, we created dashboards to help operators with real-time insights into production performance.\nWe adopted a dual-model approach: one model predicted the quality measurement, enabling continuous monitoring, while another model aimed to adjust a critical factory parameter to stabilize the product quality. Although the second model requires further refinement to achieve optimal performance, it lays the groundwork for more stable production processes.\nImpact The project provided operational insights into previously unexplored areas of the factory line. Additionaly, we revealed that product quality was more variable than anticipated. This variability had significant implications for machine tuning, which could lead to more stable production and reduced costs. The dashboards delivered included continuous predictions of quality metrics, allowing operators to monitor real-time changes rather than relying on sparse quality samples. The project ended before full-scale implementation. However, further work based on these results could help Norgips achieve more stable product quality and significant cost savings by minimizing the need for downstream corrections in the factory pipeline.\n","permalink":"https://expertanalytics.no/use_cases/gypsum_boards/","summary":"Transforming sensor data into actionable insights to reduce costs and improve product quality.","title":"Enhancing Production Processes Using Sensor Data"},{"content":"Slugging occurs when irregular surges of gas and liquid flow through pipelines, often due to variations in pressure, flow rates, or system design, causing instability in downstream processes. Intensive slugging in oil and gas production systems such as separators poses critical operational risks, including: • Gas flaring, • Oil dripping events, • Costly production stops, and • Equipment damage requiring extensive repairs. Our client, a major oil and gas operator, required a sophisticated solution to evaluate the severity of slugging as it enters a separator. The model has to handle varying levels of system knowledge, integrate seamlessly with existing operations and work in both real-time scenarios and for evaluation of large-scale simulation results of future operarions.\nApproach We developed a flexible, multi-level framework of Slugging Severity Indicators, tailored to the client’s system and data availability. The solution delivered insights at varying levels of granularity, based on the level of system instrumentation and available data:\nGeneral Indicator (Minimal System Knowledge)\nDescription: Spectral analysis of pressure and temperature variations (common measures at various points in the system) Application: Provides quick assessments with minimal system-specific input, enabling scalable deployment across diverse applications. Intermediate Indicator (Flow Rate Knowledge)\nDescription: Uses flow rates, derived from multiphase flow meters or transient flow simulations, to calculate surge volumes of slugs entering the separator. Application: ONers a more detailed evaluation, ideal for systems equipped with flow measurement tools or modeled data. System specific Indicator (Digital Twin)\nDescription: Incorporates a digital twin of the aNected components (separator and drainage control system), simulating real-world operations to predict slugging behavior. Application: Provides the most detailed analysis, supporting precise optimization of, e.g., separator performance. The framework is designed to be modular, allowing the client to adopt the level of analysis most suitable for their operational context and scale up as needed.\nImpact Once fully implemented, the Slugging Severity Indicators enable significant operational benefits, including:\nImproved Monitoring: Real-time evaluations allow for a proactive responses to slugging events, reducing the risk of unplanned production stops. Enhanced Reliability: Optimized separator and drainage system performance increase overall system reliability. Cost and Environmental Savings: Fewer flaring and oil dripping incidents minimize both operational costs and environmental impact. Future-Proofing: Improved evaluation of scenario modeling using commercial multiphase flow simulators to plan for future operations and mitigate potential risks. ","permalink":"https://expertanalytics.no/use_cases/slug_severity/","summary":"A comprehensive framework to evaluate and mitigate slugging severity in oil and gas systems, enhancing reliability, reducing costs, and supporting future operational planning.","title":"Evaluation of Slugging Severity in Oil and Gas"},{"content":"Our consultants have experience in the FinTech sector, where we helped develop customized solutions to manage risk, comply with rules and regulations, and detect and analyze potential fraudulent transactions.\nWe have experience working with end-users and managers to plan and implement technical solutions. Our consultants have worked alongside the clients\u0026rsquo; teams, developing data models, crafting ML algorithms, creating dashboards, and establishing pipelines. Moreover, we helped our clients to operationalize their data-driven ambitions, by orchestrating deployment and operation of Kubernetes clusters on cloud platforms.\n","permalink":"https://expertanalytics.no/industries/finance/","summary":"We help FinTech operationalize their data-driven needs.","title":"Finance"},{"content":"Together with Cognite and AkerBP, our consultants built a multiphase flow meter to minimise production losses for Aker BP.\nWhen multiple oil companies operate in the same field, their production pipelines are often routed to the same separator, which separates the mixed fluid into oil, gas and water streams. At this stage, ownership information is lost and to accurately identify it, measurements are done by dedicated multiphase flow meters for each field.\nThe primary challenge in this case stemmed from the Alvheim infrastructure, which limited the calibration of multiphase flow meters (MPFMs) to a single line rerouted to the Alvheim separator, causing it to exceed capacity. This necessitated production throttling or well shutdowns, resulting in significant production losses and operational inefficiencies.\nApproach To tackle the calibration challenges, Aker BP, in collaboration with Cognite and Expert Analytics, developed an innovative algorithm-based calibration method that significantly reduced the need for rerouting third-party flowlines. This approach aimed to streamline the calibration process by requiring only a single line to be rerouted, thereby alleviating the operational strain on the Alvheim separator.\nFurthermore, the new method offered the potential to eliminate rerouting altogether by leveraging planned production changes for calibration. This not only minimized production losses but also enhanced the efficiency of the calibration process, allowing Aker BP to maintain smoother operations and reduce the associated costs significantly.\nImpact The implementation of the algorithm-based calibration method had a profound impact on Aker BP by reducing deferrals by over 95%, which translated to an estimated annual savings of $3.5 Million. Additionally, the new approach minimized the strain on the Alvheim separator, allowing for more efficient operations and enabling smoother production processes, while also being scalable for future developments and tie-ins.\nhttps://www.cognite.com/en/customer-stories/dataops-oil-gas-multiphase-flow-meter-calibration\n","permalink":"https://expertanalytics.no/use_cases/flowmeter_calibration/","summary":"Using statistical analysis to reduce loss in production due to calibration routines","title":"Flowmeter Calibration"},{"content":"As a financial institution, our customer is mandated to proactively monitor risks and identify any unwanted or illegal activities. We were engaged to collaborate on the development of systems to fulfill these regulatory requirements.\nIn order to achieve swift and effective results, the team approached the challenge of implementing a transaction monitoring system by developing a set of advanced SQL queries on customer and transaction data. We collaborated closely with transaction investigators to gain insights into different risk scenarios and the measures currently implemented. It was necessary to establish new data models by combining internal and external data, creating representations that accurately depicted the domain. The transaction monitoring rules implemented by the team generated training data, which later facilitated the implementation of AI-based models to improve the accuracy of the system.\nOur customer is continuously evaluated by the financial authorities and has always met the required standards. In addition, unwanted activities are quickly detected, and users who don\u0026rsquo;t adhere to terms and conditions or Norwegian law are blacklisted and, in some instances, reported to the police. The latter has resulted in several news headlines in the last few years, creating a safer product for the users.\n","permalink":"https://expertanalytics.no/use_cases/fraud/","summary":"Development of transactions monitoring system for fraud detection in financial institution","title":"Fraud Detection"},{"content":"Expert Analytics has partnered with multinational oil and gas companies to develop innovative software solutions. These solutions leverage the power of artificial intelligence, advanced statistics, and sensor analysis. By integrating advanced software solutions into their operations, we enable these companies to streamline their workflows, improve decision-making processes, and enhance overall efficiency Our tailored solutions are designed to address the unique challenges faced by the oil and gas sector, ensuring that our clients can optimize their resources, reduce operational costs, and increase production reliability.\nMoreover, our expertise in data utilization empowers clients to unlock valuable insights from their existing data sets, transforming raw data into actionable intelligence. Through predictive analytics and real-time monitoring, we help organizations anticipate challenges, mitigate risks, and identify new opportunities for growth. This data-driven approach not only enhances operational performance but also supports strategic planning and long-term sustainability, positioning our clients at the forefront of industry innovation and competitiveness.\n","permalink":"https://expertanalytics.no/industries/oil_and_gas/","summary":"We develop smart software solutions to optimize production.","title":"Oil and Gas"},{"content":"We helped one of our clients in the Maritime Industry to develop a finite-element simulator to guide their technical choices when designing their boats.\nThe biggest challenge that our client had was that their innovative vessel design prevented the use of the classical Hydrodynamic Theory for some of the required analysis. In order to attack this problem, our consultants, together with the client\u0026rsquo;s Subject Matter Experts, derived a new mathematical theory suitable for the client vessel\u0026rsquo;s design. Then, we developed a finite-element solver to perform numerical simulations to be used as a guide for the client\u0026rsquo;s engineering choices when designing their boat.\nThe numerical simulator was developed combining industry standard best practices, such as version control, Docker, open-source packages, and user-friendliness (the intrincate details of the finite-element solver was not-exposed to the end-user).\n","permalink":"https://expertanalytics.no/use_cases/marine_vessel/","summary":"Development of a state-of-the-art simulator for guiding engineering choices in maritime industry","title":"Optimising Marine Vessel Development"},{"content":"A new method of calculating power transmission capacities for the day-ahead power market in the Nordic countries was introduced at the beginning of November 2024. This new method, called Flow-Based Market Coupling (FBMC), represents a significant change in power market operations.\nOur customer required assistance in developing a short-term power price forecasting model to meet the November deadline. The model needed to comprehensively describe production and consumption across all Nordic price areas, including capacities of international interconnector cables. A key requirement was the model’s ability to integrate daily published power line transmission capacities.\nSolution: our approach Our consultant joined the client\u0026rsquo;s internal team to:\nEstablish a comprehensive domain model to organize input data Set up the optimization problem Develop analytical tools for forecasting Document the mathematical foundations for internal reference. Our consultant\u0026rsquo;s background in the renewable power sector and experience from the colleagues at XAL provided important domain expertise.\nResults The model was completed before the November deadline and successfully deployed on the first day of Flow-Based Market Coupling.\n","permalink":"https://expertanalytics.no/use_cases/day_ahead_market/","summary":"Designing a power price forecasting model for Nordic Flow-Based Market Coupling integration.","title":"Optimizing Power Price Forecasting for Nordic Flow-Based Market Coupling"},{"content":"Expert Analytics was born in 2013 out of a shared vision between two passionate researchers, Åsmund Ødegård and Ola Skavhaug. Both had spent years working at Simula Research Laboratory, where they honed their expertise in complex simulations and cutting-edge technologies. But their ambition extended beyond academia — they wanted to take their knowledge into the real world, tackling the tough, unsolved problems faced by industries. This drive led them to start Expert Analytics, where they could bridge the gap between research and practical, high-impact solutions.\nIn the early days, the company focused on pioneering projects, like creating digital payment solutions and developing open-source simulator software for operational hydrology. As these efforts gained traction, Åsmund and Ola began expanding the team, bringing on board brilliant minds from universities—engineers and researchers with advanced degrees, who shared their passion for solving the most challenging industrial problems. Today, Expert Analytics is a diverse team of 40 experts, working with industries in hydropower, energy, industrial automation, and even oncology. The journey has been driven by a desire to push boundaries and make a real-world difference, and that mission continues to define the company’s work every day.\n","permalink":"https://expertanalytics.no/about_us/history/","summary":"Empowering the Norwegian industry since 2013.","title":"Our History"},{"content":"In our pro bono programme we seek to find solutions to interesting data science challenges together with problem owners using both classical methods and modern machine learning and data science techniques. Our target audience is composed by research institutes and groups, early start-ups, NGOs, and other (funding-limited) companies.\nThe programme is pro bono as we will use our excess capacity in it, both as part of our internal training programme aimed at furthering the professional development of our researchers, as well as part of meaningful onboarding of new employees.\nA pro bono project will usually be limited to 100 hours of work.\nSome example use cases are:\nExploratory data analysis — Do you need solutions for a broad-framed problem?\nWorkflow automation — Do you have a lot of data, but need smoother data flow?\nPredictive analysis — Can Machine Learning or Artificial Intelligence be used in your data pipeline?\nFeasibility studies — Do you have a specific hypothesis that needs investigating?\nData visualisation — How to plot complex data in a meaningful way?\nTo apply Submit a 1-2 page proposal for a pro bono project to probono@xal.no. The application should describe the problem space, available data, and desired outcome.\nSelection and initial mapping Typically, there is a two-week review period before approval or rejection. Discuss a roadmap, timeline, and deliverables for the project. Both parties will sign a simple non-disclosure agreement to protect both interests.\nProject Start-up meeting with the end-user and the expert(s) working on the project. The end user is invited to contribute to the project as they see fit.\nEnd of the project Present our results to the client. Hand over the deliverables outlined in the agreement. Social media engagement from the end user would be greatly appreciated.\n","permalink":"https://expertanalytics.no/probono/","summary":"Are you part of a small company or a research group that lacks fund? With our probono program, we can provide insights into your data and devise a simple strategy to plan your business forward!","title":"Pro Bono"},{"content":"For partners in the processing industry, we delivered intelligent digital solutions that harness the power of machine learning and advanced analytics to optimize resource usage and streamline operational processes. Our custom-built applications are designed to help clients predict, automate, and control complex workflows by leveraging real-time data and historical patterns. Through predictive modeling, our solutions provide precise insights and actionable recommendations that drive efficiencies, reduce waste, and maintain compliance with regulatory standards.\nWhether it’s forecasting demand, optimizing chemical dosing, or fine-tuning automated systems, we work closely with our clients to develop tailored applications that integrate seamlessly into existing infrastructures. Our commitment to innovation enables us to turn complex data into smarter decisions, making each process more sustainable and cost-effective. With scalable software solutions, we empower industries to proactively address their operational challenges, ensuring reliable performance and measurable impact across facilities.\n","permalink":"https://expertanalytics.no/industries/processes/","summary":"Providing smart solutions to the processing sector","title":"Processing Industry"},{"content":"A reservoir in the North Sea is challenging to produce due to solids influxes. Under certain physical conditions taking place in the formation, solids can enter the production liner and block it or erode certain elements of the pipe network. This causes important disruptions in the production of hydrocarbons, as well as economical losses due to the cost of deferred production and costly well interventions.\nSolution: our approach The client assembled an interdisciplinary team that has adopted a holistic approach to investigate -and eventually avoid- solids influxes in this challenging field. The project approaches the problem by combining applied research, use of new technologies and data science. Expert Analytics scientists participate in all three phases of the project thanks to their long research experience and versatility. Our team has developed and deployed data science models in the client’s platforms (CDF, Azure and DOF) that are used by the asset engineers to monitor the status of their wells. In addition, our scientists are assessing the client in the use of new technologies for solids detection, evaluating different vendors, and collaborating with them to propose tailored improvements to their products. Finally, we are also responsible for designing the laboratory experiments to study the properties of the flow under controlled conditions.\nResults Early solids detection allows for key decision support in the asset’s daily operations. Our tools allow for increased production (by having a real time insight into the current conditions of the wells engineers can make informed decisions on production optimization) and decreased unnecessary downtime. The client estimates the impact of the implementation of the project in a 25% reduction in losses and cost related to solids influxes.\n","permalink":"https://expertanalytics.no/use_cases/production_optimization/","summary":"An integral approach to mitigate production losses due to solid influx","title":"Production Optimization of Reservoirs in the North Sea"},{"content":"We\u0026rsquo;ve been involved in enhancing operations in the renewables sector since the foundation of Expert Analytics, specifically in the Hydropower industry.\nWe have developed advanced software solutions that streamline complex data operations, empowering power market business analysts to make informed decisions with confidence. Our innovative tools automate data workflows, ensuring efficiency and accuracy, while also enabling machine learning-based predictive maintenance for heavy assets, reducing downtime and enhancing reliability.\nFurthermore, we focus on building comprehensive dashboard solutions that visualize and contextualize time series data alongside operational metrics. This holistic view supports operations teams in monitoring performance, identifying trends, and quickly responding to anomalies. By integrating operational data with market insights, we equip hydropower companies with the intelligence needed to optimize their performance and enhance their contributions to a sustainable energy future.\n","permalink":"https://expertanalytics.no/industries/renewables/","summary":"We develop custom and general solutions for power market actors.","title":"Renewables"},{"content":"In storage hydropower systems, water is kept in a reservoir upstreams of the power plant. By opening the water ways, electricity can be produced on demand, as long as there is water in the reservoir.\nTo maximize the value of the hydro power production, knowledge about how much and when the water flows into reservoirs from the catchment areas around them is essential. This is especially important during the snow melting season.\nExpert Analytics is one of the major contributors to the open source platform Shyft!\nApproach We developed Shyft as a joint effort with Statkraft, combining data driven and physics based modeling to more accurately predict the water flow in the catchment areas. Shyft is able to fully utilize distributed weather and terrain data, to create an accurate hydrological state. This makes it possible to model snow smelting and to adapt new weather patterns, e.g., caused by climate changes. To maximize performance we implemented it as a hybrid system (C++/Python).\nImpact Shyft is released as an Open Source Software. Both the Norwegian hydro power industry and academia to use it extensively and and contribute to further development.\nResources Source code on Gitlab Documentation ","permalink":"https://expertanalytics.no/use_cases/shyft/","summary":"Shyft is an open source high performance simulator for distributed hydrological simulations.","title":"Shyft"},{"content":"Together with Cognite and AkerBP, our consultants built a Smart Monitoring tool to identify the root cause of high oil-in-water concentrations with the goal to reduce production losses.\nApproach By taking into account measurements from 200 physical sensors and 100 virtual sensors, our consultants leveraged their physical science and software skills to develop a machine-learning proxy model that could identify the most likely actor when there was a high oil-in-water concentration in the production pipeline.\nThe smart monitoring system displays near real-time data and performs calculations combining sensor values and simulator outputs,providing engineers with virtual sensors and physical properties readily available for analysis.\nImpact The developed system gives engineers the ability to troubleshoot issues related to water contamination in a single dashboard, increasing their situational awareness. This allows users to take informed actions based on the available data and solve problems faster.\nAker BP estimates an increase revenue potential of $6 millions. The system can also prevent environmental incidents.\nhttps://www.cognite.com/en/customer-stories/dataops-oil-gas-hybrid-machine-learning\n","permalink":"https://expertanalytics.no/use_cases/smart/","summary":"A recommender system to identify causes of high oil-in-water concentrations","title":"Smart Monitoring System"},{"content":"Software Engineering is the disciplined application of engineering principles to the design, development, testing, and maintenance of software. It involves a systematic approach to creating high-quality software solutions that meet specific user requirements and function reliably in real-world conditions. This field encompasses a wide range of activities, including requirements analysis, software design, coding, testing, deployment, and ongoing maintenance. Software engineers use various programming languages, tools, and methodologies to build scalable, efficient, and secure software systems. They also employ best practices such as version control, code reviews, and automated testing to ensure the software is robust and maintainable.\nOur consultants bring decades of experience in software development, having successfully worked with major companies across various industries including energy (both renewable and non-renewable), finance, telecommunications, and health. We understand the unique challenges and opportunities in your business, and leverage our deep industry knowledge to provide solutions that meet your specific needs.\nBy partnering with us, you\u0026rsquo;ll benefit from our proven track record of delivering robust, scalable, and secure software systems. Our team always employs best practices to ensure your projects are completed on time and within budget. Whether you\u0026rsquo;re looking to streamline operations, optimise systems, or implement cutting-edge technologies, our seasoned consultants are here to help you achieve your business goals and stay ahead of the competition.\n","permalink":"https://expertanalytics.no/services/software_development/","summary":"Enhance your business with our specialized software engineering consultancy, offering custom, high-caliber software solutions that foster innovation and efficiency!","title":"Software Development"},{"content":"In today’s world of everything distributed, making a system do the right thing is not always an easy task. In addition, requirements constantly evolve and business owners often like to use the newest and shiniest advancements. Developing software systems in a maintainable manner with focus on correctness becomes then a challenge. And we are up for that challenge - our software engineers usually consider the case more interesting, the more complex it gets.\nConsultants at Expert Analytics have participated in large systems development projects with many of our clients, and often take on roles as tech lead, lead developer or technical architect. And we are eager at keeping up with modern trends such as variants of Agile development, test-driven development and DevSecOps. A narrow view on the programming environment is not sufficient, but rather a holistic view that includes databases, run time and testing environments, continuous integration and deployment in addition to the development environment itself.\nWhen it comes to the actual tools to be used in a project, we try to be pretty agnostic. The preferences and competences of the client are usually the deciding factors when several tools can do the job. We are fine with that - and will adapt - as through the years we have come to understand that a tool often just decides the syntax. We contribute with a deeper understanding and the ability to solve the problems at hand - solutions can then be coded up in the selected environment. There are however situations where we understand that a tool is very wrong for the problem to be solved, and then we will speak out.\nTo get the full scope of the languages, tools, and environments that Expert Analytics can apply to your project, the individual CVs of our employees must be consulted. What we most often use is Python, Java, Go, and C++. We have worked extensively with Postgres, Azure SQL, Google Datastore and Firestore, and Cosmos DB. For continuous testing and integration we are familiar with most of the main players, and we have deployed on Google, Azure, and Amazon cloud solutions.\n","permalink":"https://expertanalytics.no/services/systems_development/","summary":"Combine software, hardware and infrastructure to deliver performance and reliability to drive your business forward!","title":"Systems Development"},{"content":"The amount of waste water that enters waste water plants is highly dependent on the time of day and of weather conditions. During periods of snow melt or heavy precipitation, the ground water level is high and foreign water enters cracks in the sewer pipes. This large variability in rate and composition makes it challenging to estimate the appropriate dose of chemicals needed to clean the waste water following regulations. Our customer was an automation company that wanted to both predict the rate of waste water on the inlet of a plant and estimate the composition of the sewer. The goal was to optimize the usage of chemicals.\nApproach To predict wastewater flow, we applied machine learning algorithms (XGBOOST, LSTM) using weather data from Norges vassdrags- og energidirektorat (NVE) and Open-Meteo APIs. By incorporating time-based features and relevant environmental variables, we improved prediction accuracy and effectively captured the cyclical patterns of water flow.\nTo support monitoring and decision-making, we developed a Streamlit application that allows users to train machine learning models and compare different models and feature sets to identify the most effective configuration. Once the optimal model is selected, the application displays real-time predictions alongside actual water flow measurements and relevant weather data. This setup enables clients to continuously evaluate model performance by comparing observed data against predictions.\nTo make the predictions understandable, we integrated SHAP (SHapley Additive exPlanations) to identify the key factors driving each forecast. Additionally, we used OpenAI’s API to generate clear verbal explanations based on SHAP values and weather data, helping plant operators understand the reasons behind the predictions.\nImpact The solution delivered accurate water flow predictions and effectively managed missing data, enhancing the reliability and flexibility of forecasting. It provided control over model configurations and included explainable AI features, allowing developers and end-users to understand the factors influencing predictions. Additionally, the real-time dashboard supported decision-making, enabling the client to manage wastewater operations with precise and actionable insights.\nThe amount of chemicals used has decreased, as reported by the client.\n","permalink":"https://expertanalytics.no/use_cases/treat_smart/","summary":"We developed an application that uses a machine learning model trained on operational and weather data to predict the rate and composition of wastewater, optimizing chemical usage and reducing costs at treatment plants.","title":"Treat Smart"}]