Development of an innovative prototype generator of a driver’s work plan in collective transport systems based on evolutionary algorithms. (Q81347): Difference between revisions
Jump to navigation
Jump to search
(Removed claim: summary (P836): On SA 42799 (2015/X), the project is commissioned by R & D to develop a new service based on the unique evolutionary algorithm of the drivers’ work plan in the collective transport management systems.The service will be based on a timetable based on input data on a whole series of variables, such as drivers, vehicles, timetable, timetable, peak hours, working conditions resulting from drivers’ working regulations, hourly rates for statutory wo...) |
(Created claim: summary (P836): SA 42799(2015/X) The object of the project is to commission R & D works aimed at developing a prototype of a new service based on unique evolutionary algorithms of driver’s work plan generator in collective transport management systems. The service will be based on scheduling on the basis of input data for a whole range of variables such as drivers, vehicles, schedules, timetables, peak hours, working conditions resulting from driver’s working r...) |
||||||||||||||
| Property / summary | |||||||||||||||
SA 42799(2015/X) The object of the project is to commission R & D works aimed at developing a prototype of a new service based on unique evolutionary algorithms of driver’s work plan generator in collective transport management systems. The service will be based on scheduling on the basis of input data for a whole range of variables such as drivers, vehicles, schedules, timetables, peak hours, working conditions resulting from driver’s working regulations, hourly rates for statutory working time and overtime, formal and legal conditions of the market in which the carrier operates and many others. The aim is to plan the work of all drivers precisely for the number of hours in accordance with the standard of m-ca, in order to achieve the minimum cost of implementation of the transport plan for the mc, and to provide minim. the amount of unscheduled transport tasks (or ensure that all transport tasks are manned every day of the month). In addition, important functionalities of the solution developed within the framework of the project are to be: even distribution of the number of working days on Saturdays and Sundays and holidays and even distribution of the number of “reserves” per day for the first and second shifts. Currently, the solutions offered on the market are imperfect because they are based on greedy algorithms. These are heuristic methods (i.e. they give approximate solutions) and the time of calculation is very long. In large data sets (large urban agglomerations), this time is unacceptable. Research problem NP – a complete solution to be solved through Algor. evolutionary analyses a much broader range of acceptable solutions to find an optimal solution. Traditional greedy algorithms make locally optimal decisions without examining the effects of these choices in subsequent steps. End user service will allow to reduce the monthly costs of the resources necessary to provide transport services (primarily the number of necessary drivers and fleet (English) | |||||||||||||||
| Property / summary: SA 42799(2015/X) The object of the project is to commission R & D works aimed at developing a prototype of a new service based on unique evolutionary algorithms of driver’s work plan generator in collective transport management systems. The service will be based on scheduling on the basis of input data for a whole range of variables such as drivers, vehicles, schedules, timetables, peak hours, working conditions resulting from driver’s working regulations, hourly rates for statutory working time and overtime, formal and legal conditions of the market in which the carrier operates and many others. The aim is to plan the work of all drivers precisely for the number of hours in accordance with the standard of m-ca, in order to achieve the minimum cost of implementation of the transport plan for the mc, and to provide minim. the amount of unscheduled transport tasks (or ensure that all transport tasks are manned every day of the month). In addition, important functionalities of the solution developed within the framework of the project are to be: even distribution of the number of working days on Saturdays and Sundays and holidays and even distribution of the number of “reserves” per day for the first and second shifts. Currently, the solutions offered on the market are imperfect because they are based on greedy algorithms. These are heuristic methods (i.e. they give approximate solutions) and the time of calculation is very long. In large data sets (large urban agglomerations), this time is unacceptable. Research problem NP – a complete solution to be solved through Algor. evolutionary analyses a much broader range of acceptable solutions to find an optimal solution. Traditional greedy algorithms make locally optimal decisions without examining the effects of these choices in subsequent steps. End user service will allow to reduce the monthly costs of the resources necessary to provide transport services (primarily the number of necessary drivers and fleet (English) / rank | |||||||||||||||
Normal rank | |||||||||||||||
| Property / summary: SA 42799(2015/X) The object of the project is to commission R & D works aimed at developing a prototype of a new service based on unique evolutionary algorithms of driver’s work plan generator in collective transport management systems. The service will be based on scheduling on the basis of input data for a whole range of variables such as drivers, vehicles, schedules, timetables, peak hours, working conditions resulting from driver’s working regulations, hourly rates for statutory working time and overtime, formal and legal conditions of the market in which the carrier operates and many others. The aim is to plan the work of all drivers precisely for the number of hours in accordance with the standard of m-ca, in order to achieve the minimum cost of implementation of the transport plan for the mc, and to provide minim. the amount of unscheduled transport tasks (or ensure that all transport tasks are manned every day of the month). In addition, important functionalities of the solution developed within the framework of the project are to be: even distribution of the number of working days on Saturdays and Sundays and holidays and even distribution of the number of “reserves” per day for the first and second shifts. Currently, the solutions offered on the market are imperfect because they are based on greedy algorithms. These are heuristic methods (i.e. they give approximate solutions) and the time of calculation is very long. In large data sets (large urban agglomerations), this time is unacceptable. Research problem NP – a complete solution to be solved through Algor. evolutionary analyses a much broader range of acceptable solutions to find an optimal solution. Traditional greedy algorithms make locally optimal decisions without examining the effects of these choices in subsequent steps. End user service will allow to reduce the monthly costs of the resources necessary to provide transport services (primarily the number of necessary drivers and fleet (English) / qualifier | |||||||||||||||
point in time: 14 October 2020
| |||||||||||||||
Revision as of 11:59, 14 October 2020
Project in Poland financed by DG Regio
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Development of an innovative prototype generator of a driver’s work plan in collective transport systems based on evolutionary algorithms. |
Project in Poland financed by DG Regio |
Statements
327,250.0 zloty
0 references
385,000.0 zloty
0 references
85.0 percent
0 references
1 October 2019
0 references
30 September 2020
0 references
ITS TECHNOLOGY – SOLVEO SP. Z O. O. SP. K.
0 references
SA 42799(2015/X) Przedmiotem projektu jest zlecenie prac B+R mających na celu opracowanie prototypu nowej usługi opartej o unikalne algorytmy ewolucyjne generatora planu pracy kierowców w systemach zarządzania transportem zbiorowym. Usługa będzie opierać się na generowaniu harmonogramu na podstawie danych wejściowych, dotyczących całego szeregu zmiennych, takich jak kierowcy, pojazdy, kalendarz przewozów, rozkład jazdy, godziny szczytu, warunki pracy wynikające z regulaminu pracy kierowców, stawki godzinowe za regulaminowy czas pracy i za nadgodziny, uwarunkowania formalno-prawne rynku, na którym działa przewoźnik i wiele innych. Celem jest zaplanowanie pracy wszystkim kierowcom dokładnie na ilość godzin zgodną z normą godzin m-ca, tak aby osiągnąć minim. koszt realizacji planu przewozów dla m-ca, oraz zapewnić minim. ilość niezaplanowanych w harmon. zadań przewozowych (lub zapewnić obsadę wszystkich zadań przewozowych w każdym dniu miesiąca). Dodatkowo istotnymi funkcjonalnościami opracowywanego w ramach projektu rozwiązania mają być: równomierny rozkład ilości dni pracy w soboty i niedziele i święta oraz równomierny rozkład ilości „rezerw” w poszczególne dni na I zmianie i II zmianie. Obecnie oferowane na rynku rozwiązania są niedoskonałe ponieważ opierają się na algorytmach zachłannych. Są to metody heurystyczne (czyli dają rozwiązania przybliżone) a czas obliczeń jest bardzo długi. W dużych zestawach danych (duże aglomeracje miejskie) czas ten jest nieakceptowalny. Problem badawczy NP - zupełny planowany do rozwiązania poprzez Algor. ewolucyjne analizują znacznie szerszy zakres dopuszczalnych rozwiązań w celu znalezienia rozwiązania optymalnego. Tradycyjne algorytmy zachłanne dokonują decyzji lokalnie optymalnych bez badania skutków tych wyborów w kolejnych krokach. Użytkownikowi końc. usługa pozwoli na ograniczenie kosztów miesięcznych jakie generują zasoby niezbędne do świadczenia usług transportowych (przede wszystkim liczba niezbędnych kierowców oraz flo (Polish)
0 references
SA 42799(2015/X) The object of the project is to commission R & D works aimed at developing a prototype of a new service based on unique evolutionary algorithms of driver’s work plan generator in collective transport management systems. The service will be based on scheduling on the basis of input data for a whole range of variables such as drivers, vehicles, schedules, timetables, peak hours, working conditions resulting from driver’s working regulations, hourly rates for statutory working time and overtime, formal and legal conditions of the market in which the carrier operates and many others. The aim is to plan the work of all drivers precisely for the number of hours in accordance with the standard of m-ca, in order to achieve the minimum cost of implementation of the transport plan for the mc, and to provide minim. the amount of unscheduled transport tasks (or ensure that all transport tasks are manned every day of the month). In addition, important functionalities of the solution developed within the framework of the project are to be: even distribution of the number of working days on Saturdays and Sundays and holidays and even distribution of the number of “reserves” per day for the first and second shifts. Currently, the solutions offered on the market are imperfect because they are based on greedy algorithms. These are heuristic methods (i.e. they give approximate solutions) and the time of calculation is very long. In large data sets (large urban agglomerations), this time is unacceptable. Research problem NP – a complete solution to be solved through Algor. evolutionary analyses a much broader range of acceptable solutions to find an optimal solution. Traditional greedy algorithms make locally optimal decisions without examining the effects of these choices in subsequent steps. End user service will allow to reduce the monthly costs of the resources necessary to provide transport services (primarily the number of necessary drivers and fleet (English)
14 October 2020
0 references
Identifiers
POIR.02.03.02-18-0014/19
0 references