Keywords: genetic algorithms, road design, horizontal and vertical alignments, CAD, digital mapping.

We introduce the software "Optimal Highway Path Automatic Search System" (OHPASS) that does the optimization simulation in the road design process which has been developed since 2000 in the NEXCO Central Research Institute. In the process of designing an expressway, a great number of evaluation items, such as economy, safety, comfort, and the negative environmental impact are used to satisfy the Road Structure Ordinance and regulations of the design points. A reduction in the cost of the project which is standard for selecting the road is demanded as well as the cost of public works. However, it is difficult to compare and consider many alignments as examining various elements in the design of the road takes huge time.

On the other hand, Digital geographical features data (DM) is being gradually introduced as using CAD becomes general in the design process of roads. With such circumstances being in order, the flow that decreases the number of processes required has accelerated as DM is used for designing. The software called "Road CAD" has the function that allows for the construction drawing and the cross-section, etc. to be automatically generated against the alignment that the designers have decided in advance. As a result, the designer's work becomes easy, but the difficulty of the work that evaluates, compares and considers many items of the alignment has not changed. Based on such a situation, it is possible to reduce the process of work greatly if the system that proposes best alignments following comparison is considered automatically. We have developed the system of making optimum simulation. Naturally, this system cannot generate all the processes automatically and some are currently done by hand. This is a system aimed at reducing work greatly when roads are designed.

In the decision of the alignment, the construction costs such as earth constructions, bridges, and tunnels are important. The reduction in the cost of construction is requested though it is necessary to consider various elements of the environment, of course. We developed the OHPASS for this purpose. This system calculates better alignments by inputting design information, such as a digital topographical map, using a standard which is particular to each road, the cost of construction unit price, etc and after calculating the cost of construction, while repeating the simulation that automatically designs the alignment many times. In the process, this system calculates each cost as well as actual cost with generating the horizontal and vertical alignment which is actually used. After using special models which express alignments, we optimize with these the use of the genetic algorithm.

In designing an expressway, it takes time to compare and consider alternative routes. Now, evaluation of various items is important, and calculation of that cost is troublesome. Therefore, if the system can propose good alignments evaluated by automatically including the cost side, the work of design can be reduced. If such a system utilizes CAD software and work to repeat the design and the evaluation can many times, therefore we can spend time on searching for better alignments and take into consideration for environment and scenery.

As we mentioned above, at present, the system evaluates alignments mainly by items, such as the cost of construction. If we compare this system with previous ones, this system has advantages that we can include more objects for consideration and comparison, so the best alignments along the items automatically can be distilled. In the future, we expect to introduce to the optimized system complex items, such as environments and scenery, and construct circumstances so the system can be used in practice.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £120 +P&P)