Dalia Čalnerytė

Combination of Image Decomposition and Artificial Intelligence to Identify Evolution of Process (VDDISTPEN)

The researcher and other members of the team solved the problem of big data decomposition, which allows achieving better results compared to classical algorithms when the evolution of the process is evaluated by summarizing the results of discrete image sequence analysis.

Numerical Models of Short-Wave Physical Behaviour in Micro and Nano Structures (FEMSHORTWAVE)

The researcher led this project. Researchers solved the problem of time and resource cost optimization. It was solved using numerical modeling tasks. By combining the results obtained at different scales (larger and smaller), the researchers modeled the physical behavior of short waves for practically acceptable computational time and memory costs.

Development of prototype of innovative champignon robotics technology

The successful development of the mushroom segment is related to automatization solutions, and so the Company’s subsidiary Baltic Champs, UAB with an external partner implemented the project “Development of prototype of innovative champignon robotics technology” co-financed by EU structural funds. The Company is using artificial intelligence and robotic systems for research and technological development to reduce the risk of infections and diseases in mushrooms, and to increase the yield and quality of the mushrooms.

Schedule Planning Algorithm for Creating the Work Schedule of Aircraft Service Personnel

An algorithm based on evolutionary algorithms is developed and further developed for staff scheduling, given a list of tasks and a set of constraints describing work rules (applied in the airport sphere). During the creation of the schedule, employee shifts are formed dynamically depending on the set of tasks, maintaining the relationship between the convenience of the work schedule for employees and the optimization of working time.

Design of Machine Learning Based Algorithm for Personnel Scheduling (APTS)

Reinforcement-based algorithm to solve personnel rostering problem was created in this project. Numerical experiments were carried out for the benchmark dataset instances published in schedulingbenchmarks.org/, which describe roster requirements (shift demand and constraints) and are popular in scientific research. Although the best-known solutions have not been obtained using this algorithm, the idea to transfer the information from previously generated schedules with different number of employees and problem horizon was confirmed. The novelty of the project is that the created algorithm enables to generate a schedule under consideration of the experience accumulated while generating other schedules under the analogous set of constraints.

Combination of Image Decomposition and Artificial Intelligence to Identify Evolution of Process (VDDISTPEN)

During the project, researchers developed a methodology determining the similarity and development of municipalities based on economic and social indicators, the results of satellite photo analysis. The obtained results can be used during political or economic decision-making, applying the good practices of municipalities to similar communities, etc.

Adapting an AI-based trip generation algorithm to work in real-world conditions

The project is under development. Researchers develop a trip generation algorithm combining geographical, price and consumer behavior data by applying stimulatory learning. The results are relevant for those who want to receive individual travel offers.