|Author(s)||Collection number||Pages||Download abstract||Download full text|
|Ковалишин О. С.||№ 2 (57)||72-81|
The high level and complexity of domestic injuries are associated with an increase in the number of road injuries and the use of modern mechanisms in domestic conditions, non-compliance with safety rules. Modern injuries have a poly structured character, which causes their severe consequences and difficulties in the different stages of rehabilitation. In this regard, in the process of rehabilitation, special attention is given to the issue of individually oriented rehabilitation services, since at present plans for rehabilitation therapy are very limited to the interests of patients themselves, which directly affects the quality of treatment.
In order to solve the problem, it is necessary to take into account the set of restrictions imposed on the process — the need to consolidate schedules of the medical equipment usage, the availability of qualified medical personnel for each specific procedure at a certain time, the availability of off-the-shelf equipment, etc.
In fact, to solve this problem, it is necessary to construct a full functioning schedule of the medical institution, which is superimposed on the set of hard restrictions — the conditions that must necessarily be fulfilled to ensure the correctness of the schedule, and soft requirements — the implementation of which is desirable — the actual interests of patients. The following methods are used to solve the problem of constructing a schedule for medical institutions: full selection, method of branches and boundaries, logical programming with constraints, graph coloring, simulated annealing, simulation modeling, genetic algorithm, etc.
Regarding other methods, genetic algorithms have several advantages, such as: simultaneous use of several points of the search space, the use of probabilistic and deterministic rules for the transition between solutions, the absence of the need to use an external extra data and operations. Therefore, it has been suggested to use genetic algorithms to solve the problem of multi-criteria optimization of rehabilitation therapy plans. To assess the progress of optimization, a mechanism for evaluating alternatives generated during the work of evolution operators has been constructed. A fuzzy logic controller, the T-Controller, has several advantages over classical methods: high accuracy of computations, zero methodological error, high speed of operation.
Indicators based on fines imposed for violation of a certain criterion are used as input arguments of the controller. For the organization of the optimization process, the mechanisms of population development — crossover and mutation operators have been developed.
Keywords: multicriteria optimization, genetic algorithms, fuzzy logic, rehabilitation therapy.