JultikaUniversity of Oulu repository

The number of people with dementia is growing as people live longer. The old-age dependency ratio is predicted to increase in Europe. Institualization of the seniors is an undesirable solution in respect of quality of life and economics. Cooking is an important part of independent living, and it could be supported with assistive technologies, such as prompting systems to increase the independence of seniors. Because of the varying and individualistic effects of dementia, different seniors require tailored assistance from the prompting system. This design science research is part of ASTS project that designs new technologies for assisting seniors with dementia to remain independent. The aim of this thesis is to develop a method that can be used to measure the level of difficulty in cooking tasks for individual seniors when the aim is to find an optimal task-person combination. This aim is accomplished by researching existing research to develop a test procedure. The test can be used to model the actions of the focus group, the difficulty of the task and the level of assistance required. The results are then used to create an individual action model that can be used to compare differences between individuals and to optimize a prompting system to assists seniors in cooking activities for the client. According the requirements of the client, a simple oven porridge recipe was chosen for the tests. Using existing theories the recipe was redesigned into workflow process model with very specific steps on how to cook porridge. The process model was validated with two pilot tests where a prompting system was imitated by verbal prompts. The actual tests were conducted with two researchers and three participants in a kitchen that was available for the participants. The tests were analyzed by adapting existing studies and methods for evaluating cognitive impairment in daily tasks. The resulting action models illustrate the workflow of senior with dementia and how to measure difficulty in kitchen tasks. The method design process was evaluated with the client, and the results of the study were compared to existing research. The designed method for measuring difficulty in kitchen tasks has five steps. The first step is to redesign a recipe into workflow process model that is built out of take, give, move and alter actions. The second step is to design a test layout for the chosen task. Alternatively, the process model and layout proposed in this study can be used. The third step is to conduct the test by giving the participant different level prompts. In the fourth step, errors that occurred are identified. The fifth step is to create the action models by calculating probabilities of each error occurring.