Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!quanta.eng.ohio-state.edu!rcgl1.eng.ohio-state.edu!CHETAN From: CHETAN@rcgl1.eng.ohio-state.edu (Chetan Rastogi) Newsgroups: comp.theory Subject: Help!! Message-ID: <6291@quanta.eng.ohio-state.edu> Date: 5 Dec 90 00:33:45 GMT Sender: news@quanta.eng.ohio-state.edu Organization: Ohio State University Lines: 37 Nntp-Posting-Host: rcgl1.eng.ohio-state.edu PARDON ME IF THIS NOT THE RIGHT NEWSGROUP ... Hi, I would like your input on the particular problem oultined below. I have been wondering about this problem for some time now, and this really isn't my area of work. ( I am a Mechanical Engineer) The problem is to do with algorithmically traversing distance between two specified points which are not connected directly. Picture an airline chart or a shipping route. There are, say, N vertices (destinations), and say, N+1 paths between these vertices (what i am saying is that each vertex has a path to itself). Problem: 1) If I want to get from vertex A to vertex B, how do I find a way to get there in shortest no. of hops. ( The traveller wants as few changeovers as necessary - at this point the length of the path is immaterial). Is there an algorithmic way in which I can get the shortest (in terms of no. of changeovers) path(s), and classify all paths in an ascending/descending order. 2) Attach a penalty function relating the distance of each leg. E-mail answers/references preferred, as I am not a regular reader of news. Pseudo code or regular english is fine. If any computer professionals out there want to supply code as the algorithm, thats fine too, but please bear in mind that i am not a regular programmer, so please COMMENT intensively. Thanks a lot in anticipation Chetan Rastogi Dept. Of Mechanical Engineering The Ohio State University Brought to you by Super Global Mega Corp .com