memetic algorithm
در نشریات گروه پزشکی-
Background and Objectives
The pharmaceutical supply chain should provide medicines in the right quantity, with the acceptable quality, to the right place and customers, at the right time and with optimum cost. Pharmaceutical companies, the most important player in the pharmaceutical supply chain, have a crucial role. An integrated approach in production planning and scheduling in pharmaceutical companies typically optimizes several consecutive stages in the pharmaceutical supply chain because these companies, are both linked to suppliers and distributors/end users. Hospitals are the most crucial end-user among the end-users because of the necessity of a massive amount of drugs to render service to a large number of patients in hospitals. Intravenous fluids and irrigation solutions help patients maintain the optimal internal temperature and increase comfort in hospitals during procedures like surgery or recovery. In this work, simultaneous production planning and scheduling in a real-world application, a production line of intravenous fluids and irrigation solutions at Darou Pakhsh Pharmaceutical Manufacturing Company is addressed.
MethodsA novel mixed-integer linear programming model is formulated for multi-period simultaneous production planning and scheduling. Since the problem is NP-hard in the strong sense, a memetic algorithm is proposed that reduces the computational effort of the problem. The chromosome representation is based on a permutation matrix, and a new algorithm is developed to construct a complete schedule from the permutation matrix through the planning horizon.
Results40 problems were investigated, containing 36 randomly generated instances and four real problems according to the data within the last two years. The generated instances were divided into small-sized, medium-sized, and large-sized instances. Among the 36 instances, 22 instances were optimally solved by both the exact method and the proposed memetic algorithm. The average gap for small-sized, medium-sized, and large-sized instances are respectively 0.00%, -1.15%, and -51.38%, indicating as the size of instances grows, the gap becomes considerable. The exact method could not reach an optimal solution for four real instances. The running time for real instances is expanded to 8 hours. The results revealed that the proposed memetic algorithm significantly outperformed the exact method in obtaining better solutions for real instances.
ConclusionsThe computational results showed that the proposed memetic algorithm obtained optimal solutions on all the instances solved optimally by the exact method. It outperformed the exact method in other problems. This outperformance becomes more evident as the size of instances grows.
Keywords: Simultaneous production planning, scheduling, production line of I.V. fluids, irrigation solutions, mixed integer linear programming, Memetic algorithm, Darou Pakhsh Pharmaceutical Manufacturing Company -
مقدمهالگوریتم های فرا ابتکاری و ترکیبی از توانمندی بالایی در مدل سازی مسائل پزشکی برخوردارند. در این مطالعه از شبکه عصبی به منظور پیش بینی ابتلا به دیابت در میان افراد مستعد دیابت استفاده گردید.روش کارپژوهش حاضر از نوع کاربردی و جامعه ی هدف آن متشکل از 545 فرد بیمار و سالم از مرکز دیابت دانشگاه علوم پزشکی همدان جمع آوری گردید جهت پیش بینی بیماری دیابت استفاده شده است. در این مطالعه از الگوریتم ممتیک که تلفیقی است از الگوریتم ژنتیک و الگوریتم جستجوی محلی است، به منظور به روز رسانی وزن های شبکه عصبی و توسعه دقت شبکه عصبی استفاده شده است.یافته هابررسی اولیه نشان داد که دقت شبکه عصبی، 88درصد، می باشد. بعد از بروز رسانی وزن ها با الگوریتم ممتیک دقت آن به 2/93درصد افزایش یافت. برای مدل پیشنهادی به ترتیب حساسیت، ویژگی، ارزش اخباری مثبت، ارزش اخباری منفی، مساحت زیر منحنی 2/96، 4/92، 8/93، 3/95، 958/0 برای مدل الگوریتم ژنتیک، 98، 8/84، 6/88، 2/98، 952/0 و برای مدل رگرسیون لجستیک، 6/95، 5/84، 7/94، 0/87، 916/0 به دست آمد.نتیجه گیریبر اساس یافته های این پژوهش، مدل های شبکه های عصبی در مقایسه با مدل رگرسیون از میزان خطای کمتری در تشخیص بیماری بر اساس متغیرهای فردی و سبک زندگی برخوردارند. یافته های این مطالعه می تواند به برنامه ریزان و ارائه کنندگان خدمات سلامت در برنامه های غربالگری و تشخیص به موقع بیماری دیابت کمک می نماید.کلید واژگان: دیابت، تکنیک پشتیبان تصمیم گیری، شبکه عصبی، الگوریتم ژنتیک، الگوریتم ممتیکIntroductionMeta-heuristic and combined algorithms have a great capability in modelling medical decision making. This study used neural networks in order to predict Type 2 Diabetes (T2D) among high risk individuals.MethodsThis study was an applied research. Data from 545 individuals (diabetic and non-diabetic), in Diabetes Clinic of Hamedan University of Medical Sciences, were used to develop predictive diabetes models. Memetic algorithms which are a combination of genetic algorithm (GA), local search algorithm, and neural networks were applied to update weights and improve predictive accuracy of neural network models. In the first step, optimum parameters for neural networks such as momentum rate, transfer functions, and error functions were examined through trial and error and other studies.ResultsThe preliminary analysis showed that the accuracy of neural networks was 88 percent. The use of memetic algorithm improved its accuracy to 93.2 percent. Among models, regression model had the least accuracy. For the memetic algorithm model the amount of sensitivity, specificity, positive predictive value, negative predictive value, and area under the curve were 96.2, 95.3, 93.8, 92.4, and 0.958, respectively. These parameters for GA were 98.0, 84.8, 88.6, 98.2, and 0.952 and for the logistic regression model were 95.6, 84.5, 94.7, 87.0, and 0.916, respectively.ConclusionsModels developed by neural networks have a higher predictive accuracy than the regression model. The results of this study can contribute to risk management and planning of health services by providing healthcare decision makers with more accurate predictive models based on clinical and life style characteristics of individuals.Keywords: Diabetes, Decision Support Techniques, Neural network, Genetic Algorithms, Memetic algorithm
- نتایج بر اساس تاریخ انتشار مرتب شدهاند.
- کلیدواژه مورد نظر شما تنها در فیلد کلیدواژگان مقالات جستجو شدهاست. به منظور حذف نتایج غیر مرتبط، جستجو تنها در مقالات مجلاتی انجام شده که با مجله ماخذ هم موضوع هستند.
- در صورتی که میخواهید جستجو را در همه موضوعات و با شرایط دیگر تکرار کنید به صفحه جستجوی پیشرفته مجلات مراجعه کنید.