<< | Home Page | Latest News >>
Van: "Daria Madjidian" 21 februari 2007 Aan: Demotech
Onderwerp: Some thaughts
Hi Reinder,
Hope everything is good. I have some thaughts I would like to share. This letter is very unorganized (3.30 am here) so I hope you can extract some info. If not, give me your landline number and I'll clarify over the phone.
About the transportation capacity: We could use Mexico City as an example. I emailed a Mexican embassy in Canada (copied you on it) but I don't expect an answer, so I'll call them. Also I found this link that haven't been useful yet http://portal.sct.gob.mx/SctPortal/. Maybe it can direct us further though.
I got the max # of passenger/(h and direction) to 2000 with the numbers you gave me: Car frequency * 0.8 = 1/time between cars * 0.8 = 15000km/h / 6m * 0.8 = 2000. I did a quick comparision to Montreal's subway system. During rush hour there are trains every 3 min and each train takes 1500 pass. That's 30.000 pass/h in each dir. To be able to match that, Izigo would need about 15 lines for each dir. With 400 m between them the lines would cover 6 km2, which is probably a good estimate for a large city centre. Of course we're comparing to a well functioning infra structure during rush hour. Most 3rd world cities probably can't afford an infrastruture like that, so if Izigo is more economic it wold probably be a good option. But then again, we have to consider that the Metro can handle larger disturbances than Izigo (big pipe vs several small pipes analogy).
I am hoping to find estimates soon. The best thing would probably be to compare it to the bus system of a large city without subway (excludes Mexico city). Unfortunately I didn't work on that level (planning of capacity) during my (MSc) thesis. Our issue was stability. However I think it looks good and I'm optimistic.
About routing: I think I understand the chicken mesh thing. I tried that with a few examples using Dijkstra's algorithm and re-calculation at each node. And it's possible for the car to get stuck (being sent back and forth). I think it can be fixed by making the algorithm less greedy, possibly by introducing hysteresis. But before I try I'm going to consult my friend in networking.
I'll let you know once I know more. I hope to get a scanner soon so I can illustrate my thaughts better.
Enjoy the morning, Daria
Begin doorgestuurd bericht:
Van: "Daria Madjidian" 21 februari 2007 Aan: nstn0281@fox.nstn.ca, mexgo3@inforam.net
Onderwerp: Need your help on finding information
Dear Sir/Madam,
I hope you are doing well. I also hope you can help me. I am currently working on a new mass transportation system for a technical NGO (http://www.demotech.org/ ). The system is meant to serve as means of transportation in large cities. At this stage we need to know the amount of people that the system needs to support. Since Mexico City is the world's largest city as far as I know, I would like to use it as a reference. I am trying to find the following information:
Estimates on the number of people/h that travel (average and rush hour).
How many travelers/h the public transport system supports.
How many people travel by car each day?
I would really appreciate your help. If you do not have the information, I would be very grateful if you could direct me to the right person/department.
Thank you very much for your time,
Daria Madjidian
Van: "Daria Madjidian" 20 februari 2007 Aan: Demotech
Onderwerp: Antw.: Community cooperation
Hi Reinder,
Your thaught is very good. I think as we go and we get people interested it will spread. If we just keep an open mind we'll find people. I'm back in Montreal now and hope to get some work done this afternoon!
I gonna get back to you with questions on your drawings, as soon as I have formulated them. I also talked to my friend here. He is in networking and he should be able to help on Izigo. He said he was interested and we'll talk about it more. I'm gonna give him the Wiki page and your email too.
Enjoy the evening and hope the blues is good tonight too :)
Daria
P.S. I speak Farsi because my father i sfrom Iran and I grew up speaking it. Similarly I speak Greek because my mother is from Greece.
Van: Demotech, dd 19 februari 2007 4:19:40 GMT+01:00, Aan: Daria Madjidian
Subject: Community cooperation
Daria, hello!
This night is just to nice to sleep. There is tantalizing blues on the radio and it feels good to have made some clear progress these days. Bram does good work in Guatemala, I myself will soon start of for an other project area, still to be decided upon. I really like to be partnering with you on this Izigo project, that because of that suddenly is a realistic option again. Extra is that here in Maastricht at last some support in management and finance emerges. I have done it for 20 years without! Meaning crucial cooperation can be organized again. Some administrative support, some handy person for help on the prototypes. It would make such a difference.
What would fit in is getting design communities started. Something as you mentioned, contacting your friends that could help as well. Then looking very good at the example of Linux and how that program was developed by a great many smart people working effectively together.
Would you have an idea how to organize that? Would you know sources for this type of info? How good are your connections with your university in Sweden? Developing hardware should be a different art then the well practiced development of software. I can not see that much difference between the two in regard of design.
Ok, this was just one extra thought that lingers around this night. Another: how come you speak fluent Parsi?
Have a nice Monday!
Reinder
Van: Demotech, 19 februari 2007
Onderwerp: Doorst.: A bit of work sunday?
Look for capacity figures. I have no idea of the amount of traffic this system has to cope with. How many people out of an underground station every hour? How many of these station per square kilometer? This number the same for big and smaller cities? What would public transport (including privately run lines) in poor countries be in numbers? Calcutta? Mexico City? Your study for the flat Paternoster (your phd-object) must have given you an impression.
Transport density per line per hour in one direction, an estimate can be made according to the following assumptions: - All lines are used in one direction only. - People transported along the Izigo lines could be 0,8 passenger per cabin (empty cabins relocate themselves automatically to places of demand). - Best performance of the lines would be at a distance between cabins of two cabins length. This allows a cabin from a side line to slide in with minimal speed reduction. After entry of an extra cabin on the line, speed correction of all cabins make them resume the 1 cabin in 3 cabin-length units as an optimum. Electronic sensors are expected to be able to make this possible. - Speed of the cabin could be 15 km/hour. With this speed running into a cabin that stands still may work it shock absorbers reduce the speed from 15 km/h to standstill over a length of 1 meter, being double the impact length of an air filled rubber buffer. This would be the worst possible accident. Normal bumping would have to be much softer. - Length of the cabin including the buffers could be 2 meters From this the capacity can be calculated: 15.000 m/hour x 6 m / 3 x 2m/cabin x 0,8 p/cabin = 2400 passengers per hour per line. - The grid of lines covering the centre of the city could be 400 meters wide, or every 400 meters perpendicular on the direction of the line, the next line in the same direction can be found and so on. This would give a flow of people in one direction of 1000 m/400 m x 2400 passengers per hour = 6000 passengers per hour per 1 km cross section.
Is that little or much? Hope you have an idea!
I'll do a rough draft of the chicken mesh picture. Hope it is enough to fuel your imagination. I hope to send it to you in one hour's time.
Van: "Daria Madjidian" 18 februari 2007 Aan: Demotech
Onderwerp: Antw.: A bit of work sunday?
Hey Reinder,
I think I have a good idea of what you mean, but unfortunately the chickenmesh dosen't help me right now. I'm gonna see if there is info about it though. And please also clarify "intervwoven layers". I thought you would have separate lines for every direction on each branch. Or is the plan to use each line for dual directions?
Right now I'm in Toronto and I'm visiting friends and family. Otherwise, I live in Montreal. It has about 4 million people and I leave in the best part of town :) And you're right, Montreal is covered with a lot of snow right now. How's Holland?
Best, Daria
On 2/17/07, Demotech <info@demotech.org> wrote:
Daria, hello!
Funny to have spoken to you. The line was real bad. I had difficulty
with getting the message. Anyway, my first task is to line out the
working of the grid. The need of explaining it to you made me think
of chicken mesh as the best metafor. That is for the North-South and
East-West movements. But the opposite direction also ask for
chikenmesh, while the two layers will also have to be interwoven. I
hope your mathimatical mind will solve this puzzle.
I hope to do something about it tomorrow Sunday.
How do you spend your weekends? Canada is covered bij snow now. Do
you stay in a town or countryside?
Till tomorrow, greetings, Reinder
Van: Demotech, 19 februari 2007 Aan: Daria Madjidian <d.madjidian@gmail.com>
Onderwerp: Doorst.: Images of traffic flow
Daria, I promised to send an indication of the chicken mesh pattern. That is about the same as the loop interaction, possible more effective. Mesh layers on top of each other can interact, when such top layers would interconnect larger sections. All this type of refinement will come later. First we should know if the transport capacity is sufficient. I hope my calculation makes sense.
Greetings again, Reinder
Van: Demotech, 19 februari 2007 Aan: Daria Madjidian
Onderwerp: Images of traffic flow
Daria, hello again!
Here they are, two rough sketches. Hope you can follow the idea. #1 is an indication of the principal setup. A central loop with satelite loops. A station is indicated in every loop.
Greetings, Reinder
DSCF0010.JPG
Van: Demotech , 19 februari 2007 Aan: "Daria Madjidian"
Onderwerp: Antw.: A bit of work sunday?
Daria, hello!
I am working on it right now. Hope indeed you find that info. Also look for capacity figures. I have no idea of the amount of traffic this system has to cope with. How many people out of an underground station every hour? How many of these station per square kilometer? This number the same for big and smaller cities? What would public transport (including privately run lines) in poor countries be in numbers? Calcutta? Mexico City? Your study for the flat Paternoster (your phd-object) must have given you an impression.
Transport density per line per hour in one direction, an estimate can be made according to the following assumptions:
From this the capacity can be calculated: 15.000 m/hour x 6 m / 3 x 2m/cabin x 0,8 p/cabin = 2400 passengers per hour per line.
Is that little or much? Hope you have an idea!
I'll do a rough draft of the chicken mesh picture. Hope it is enough to fuel your imagination. I hope to send it to you in one hour's time.
Greetings, Reinder
Op 18-feb-2007, om 23:57 heeft Daria Madjidian het volgende geschreven: . Hey Reinder, come hI think I have a good idea of what you mean, but unfortunately the chickenmesh dosen't help me right now. I'm gonna see if there is info about it though. And please also clarify "intervwoven layers". I thought you would have separate lines for every direction on each branch. Or is the plan to use each line for dual directions?
Right now I'm in Toronto and I'm visiting friends and family. Otherwise, I live in Montreal. It has about 4 million people and I leave in the best part of town :) And you're right, Montreal is covered with a lot of snow right now. How's Holland?
Best, Daria
On 2/17/07, Demotech <info@demotech.org> wrote: Daria, hello! Funny to have spoken to you. The line was real bad. I had difficulty with getting the message. Anyway, my first task is to line out the working of the grid. The need of explaining it to you made me think of chicken mesh as the best metafor. That is for the North-South and East-West movements. But the opposite direction also ask for chikenmesh, while the two layers will also have to be interwoven. I hope your mathimatical mind will solve this puzzle. I hope to do something about it tomorrow Sunday. How do you spend your weekends? Canada is covered bij snow now. Do you stay in a town or countryside?
Daria, hello, Real short, as it is too late here. Tomorrow leaves Bram to Guatemala for field work on the hydraulic ram (see http://www.youtube.com/watch?v=f4ngVxNF7Uw ) and we discussed last minute items.
For contacting me any time tomorrow try to use Skype. My Skype name is reindervt Please try to use it, better then the fixed line as I do not have a fixed line of my own anymore as a mobile phone seems more useful. But notify me you want to Skype by calling my present mobile number 0031 6 167 620 51 I'll then open my computer and skype, so we can talk. And mail me your Skype name as well.
Sorry, I am nowhere in mathematics, I have no idea about routing algorithms, hardly know how to spell that word. But I do have an idea how the Izigo circuits could work. It will be your modeling that will show that it indeed can work and what capacity the system will have. I am a bit shaky about this truth coming out. Could well be that I have been far to optimistic and that I forgot a few very important aspects. So I really look forward to your findings.
To my feel disturbance propagation is not an issue here, as there is no stiff (hard) interconnection between the node loops or lines.
I made a section on the Demotech Wiki-section with the name "Izigo Development Community". You find it at http://www.demotech.org/pmwiki/pmwiki.php?n=Izigo.HomePage . To keep some order in all the data we will exchange, using these Wiki pages may be most optimal. I transferred out correspondence to it and also the links to the Izigo pages. Working with Wiki is easy enough. Please try it out, it is very convenient.
In the link to a page with the number sub=143 there is the chapter How the Izigo Transportation system may function and look like There I try to explain how the routing of the cabins could be done.
Daria, that's it for today.
Hope very much to speak to you tomorrow!
Reinder
Van: "Daria Madjidian", 14 februari 2007 Aan: Demotech
Onderwerp: Antw.: Izigo!
Hi Reinder,
That's good information. I need a bit more info about the routing (highlighted in orange below). I need to get up to date on routing algorithms and I'm wondering if you thought about any algorithms (Dijkstra, etc)? Just so I know where to start...
Vehicle platoons is a chain of automated cars following each other. One lead car and the rest are followers. By disturbance propagation I mean: if one car is disturbed, it's neighbours are disturbed, and in turn their neighbours and so on... If there is no way of breaking down the system into smaller sections (and if the load is significant) then eventually the disturbance will affect the car that was initially disturbed. Think of it as a system of masses connected with springs. Then disturb one of the masses.....
Phone/Skype is a good idea and we can talk about this more on the phone. We should probably speak soon, so we can get exchange info in a faster way. How id your schedule tomorrow? Friday-Monday I'll be in Toronto and it would be nice to have material to work.
I really like your idea, and I'm very happy that we can work together.
Best, Daria
On 2/14/07, Demotech < Info@demotech.org> wrote:
Daria, hello again!
Just a quick and short answer as I need more time then available at this very moment for being complete.
First of all I am very happy and exited that you like the Izigo idea and want to work on it. It kind of hurt that I had to leave this design initiative alone for so long. The Izigo idea addresses a real huge problem all larger cities face. If there is a little chance to find a way out, it should be worked on. Ok, there we go:
Indeed I did a lot of thinking on this project. I'll write down some of the ideas generated
I just now looked at the link you enclosed in your mail. Quiet impressive! It indicates to me that you are fully aware of what is involved in urban mobility systems and that you have the capacity to dig into it far deeper then I myself could ever go. This is good news!
There is so much more to tell about considerations and ideas. So we have to set up a kind of station where our information can go to be worked on. For the time being I want to make the same Wiki section as I made for Olle Wellin. I'll do this soonest. Discussing by phone (fixed line) is also cheap, about 1 ct/min from the Netherlands to Canada. Best may be to use Skype as then we can also show to each other sketches we made. And it is fun to see each other life!
I look forward to all that is going to happen in the Izigo project. It is great that you joined its development, thank you for that!
Good luck!
Reinder
Op 14-feb-2007 heeft Daria Madjidian het volgende geschreven:
Hi Reinder,
Thank you very much for your kind e-mail. I find the idea fascinating, and I've started thinking about it a bit. From what I understood from your e-mail, it is at the embryo stage. Although the information in the link suggests that a considerable amount of thought has been put into the project. As you said, the first problem would be to find a routing strategy. From the web-site I understood that the plan is to re-calculate optimal route at each node. Since the load (# of cabins on each line) always varies, I'm thinking that it could lead to cabins being sent back and forth. In order for me to be of best use and know to what problem I should devote my energy, I would like to know:
How much work have been done in developing the routing strategy?
Is the plan for the cars to have variable speed? (This requires control similar to vehicle platoons)
Are there any ideas on instrumentation yet? I.e. detectors, sensors etc for position and speed information?
How are the cabins driven?
That's all I can think of for now:) I see a major (and fun) problem being disturbance propagation in the system. Variable speed, would increase the propagation. Also many cabins without a way to break the system down into sections makes propagation problem more difficult. A way of breaking it down could be to use a some form of traffic light system. I think this problem is a bit similar to the one I worked on for my thesis. Each cabin must take info about other cabins into account. See chapter 3 in http://www.control.lth.se/database/publications/article.pike?artkey=5748 .
I read your manifesto and I like your philosophy. I'm very happy that there are people like you out there. I also like the idea of focusing on students and young professionals to do most of the work. I have a few people in mind and I will try to get them involved as well.
Thanks once again Reinder. I look forward to hearing from you and getting started.
Best regards, Daria
On 2/13/07, Demotech <Info@demotech.org> wrote:
Daria, hello!
Good to hear again from you! I was surprised that you hopped to another country and continent. But not after looking at your CV and becoming aware you like to be on the move. ou really make me glad with your offer for considerable involvement. What you offer is not little! I hope that Demotech can inspire you enough to make you enjoy it.
Indeed -after reading your CV- I have an idea of a Demotech design initiative that you may like to work on. Even it seems to me that you have precisely the expertise this project needs at the early state it is still in.
The word 'design initiative' is just right: it is not a design, but it certainly is an initiative. It has a name: Izigo (easy+go). The idea is that all the growing mega-cities need a better system for public transport. And that this type of transport could be created by the strange mix of the technology of the fair with the technology of Internet.
The mix of the little bumping electric carts roughly driven by daring youth at the fair, being set up over night to disappear only a few days later, the mix of this technology run by the lowest educated as well as paid level of technical specialists, the mix of this technology with the packet switching technology of Internet, that may shape the Izigo system.
You will find illustrations and a description of Izigo at http://www.demotech.org/d-design/designA.html?d=34
Please be so kind as to read this as well as the description you will find at http://www.demotech.org/design/designB.html?sub=143
At this design stage we have to find out how the Izigo system will work as a transport system, taking for granted that technical problems are solved. These technical problems are not solved at all, but there comes in the hope that IF the system can work according to the robust technology of the fair and as I estimate that these technical problems are of this level and of the level of the banana transport system as found in Costa Rica, then there is just a technical puzzle to be solved.
But it is the modeling of traffic streams in different directions during the day, with different cart speeds, different grid size, different ways of the merging of cart lines that shows if and how
this public transport system can be effective. The outcome of this study gives the motivation and the contours for further research.
Also I read in your CV that you were teaching at an university and have given guidance to students. It is my hope that university students will push universities into research directed at a sustainable and just world. Demotech gets support from a student association called the 'Workforce'. When we started this group, we wrote 'The Manifesto', see http://www.demotech.org/pmwiki/pmwiki.php/Design/DemotechManifesto . There are already many universities where students and teachers work that way, possibly MIT and the Humbold University (look up Appropedia). Can you imagine that you could inspire and guide students in the work on the Izigo according to these ideas?
I realize that getting results with this project is a real big job and much to much for one person. So I hope to find a way to increase your capacity in a way you really like and find productive. Here again modeling has to be done, though this time in the social realm.
Daria, thanks again for your offer. If the above does not suit you for whatever reason, let me know and we will construct something else that you may like. Have you ever heard of the Social Construction of Technology or SCOT? It is a sociological theory posed by Wiebe Bijker at the Maastricht University in the Netherlands, the same location where my 'Lab' is situated in a deserted factory. I hope to involve him too, as his type of thinking suits me very well.
But first tell me what you think of all this.
Receive my kind and happy regards,
Reinder
Op 13-feb-2007, om 20:25 heeft Daria Madjidian het volgende geschreven:
Hi Reinder, Hope you are well and hope 2006 was a good year for you. I am contacting you once again since I want to make a contribution. I am sorry that I couldn't be of use last year. I relocated to Montreal, Canada a month ago and thus the last few months in Sweden were very busy. Now everything has calmed down, and I am pretty settled.
Please tell me if you need any help. I have a MSc in electrical engineering and I am specialized in automatic control. My specialization is mathematical modeling and simulation, control algorithms design, stability analysis and optimization. I don't mind trying new things, but these fields are probably were you can make best use of me (last time I tried to look at Olle's Stirling Engine but I didn't have time to make it through the learning stage). I've attached a CV that explains my background in more detail. If you want you can send me a list of projects and areas you need help in and I can tell you what I can do.
Currently I can put in at least 10 h a week and maybe more. But probably not more than 20 hours a week. Please don't hesitate to contact me if you have any questions and concerns. And once again, sorry for not being of more help in the past.
Looking forward to hearing from you.
Best regards, Daria
Attached CV - Daria Madjidian.pdf