Akademik unvan temelli tercihlere ve sapmalara dayalý ders çizelgeleme modeli-endüstri mühendisliði bölümü örneði

Ders programý çizelgeleme problemi, üniversiteler ve benzeri eðitim kurumlarýnda yaygýn olarak karþýlaþýlan zaman çizelgeleme problemlerinden birisidir. Çözümü, yoðun iþ gücü ve kaynak gerektirmekte olup birçok eðitim kurumunda çizelgeleme problemi halen elle yapýlmakta ve çok zaman kaybettirmektedir. Ders programý çizelgeleme problemi, eðitim kurumlarýna özgü kýsýt yapýlarý dikkate alýnarak, derslerin uygun zaman dilimlerine atanmasýný konu almaktadýr. Kýsýt yapýlarý, eðitim kurumunun özel kurallarý, kapasite kýsýtlamalarý, yasal düzenlemeler, öðretim elemanlarý ve öðrencilerin tercihleri gibi farklý faktörlere baðlý olabilir. Bu çalýþmada, öðretim elamaný tercihleri doðrultusunda iki yeni 0-1 tam sayýlý programlama modeli önerilmiþ, bir devlet üniversitesinde yapýlan uygulama ile modeller test edilmiþtir. Önerilen ilk modelde, öðretim elemanlarýnýn memnuniyetleri en büyüklenmek istenirken, istenmeyen ders çakýþmasý da en küçüklenmek istenmektedir. Ýkinci modelde ise, ilk model amaçlarýna ilave olarak ayný unvana sahip öðretim elemanlarýnýn memnuniyet deðerleri arasýndaki farkýn da en küçüklenmesi amaçlanmýþtýr. Önerilen modeller, GAMS 23.8.2 yazýlýmý ile çözülmüþ, karar vericiler için uygun olabilecek seçenekler sunulmuþtur.

A course scheduling model based on academic title-based preferences and deviations- a case study of industrial engineering department

Course timetabling problem is one of the most common time scheduling problem that frequently encountered in universities and whole education institutions. To eliminate this problem requires notable amount of labour and supply and in several education institutions, this scheduling is done manually but it doesn’t occur as desired and causes to loss of time. Course timetabling problem is the assignment of courses to appropriate time periods, considering the constraint structures specific to the educational institutions. Constraints may depend on different factors such as the specific rules of the educational institution, capacity constraints, legal regulations, faculty and students' preferences. In this study, two new 0-1 integer programming models were proposed by considering preferences of the faculty members, and the models were tested with the application in a state university. In first proposed model, while the total gladness value is desired to be maximized by considering the time-slot preferences of lecturers, the conflict of courses is tried to be minimized. In the second proposed model, the difference in gladness value between lecturers are aimed to be minimized in addition to first model objectives. The proposed models were solved with GAMS 23.8.2 software, and alternatives that could be suitable for decision makers were presented.