PRAXITÈLE UPDATE - DECEMBER 1998
(Draft in process. Awaiting review and comment)

To get us going on this the best introduction you will have on this will be
on the Praxitèle site itself at www-rocq.inria.fr/praxitele/.  Of course
these materials are all in French; if you require English language backup we
would refer you to our latest Casebook on the CarSharing Consortium site at
www.ecoplan.org/carshare/cs_index.htm.  In the meantime as a quick reminder,
I attach below a short English language summary description of the
technology side of the project taken from the Praxitèle site.

Against this solid base, let me see if I can now quickly share with you the
high points that came out of our half day site visit.  I should add that I
was lucky to have the company of Susan Shaheen of the Institute of
Transportation Studies-Davis ([log in to unmask]) who is extremely
knowledgeable and these matters and, as most of you know, is also the
project manager for the new Car Link project which is about to start up next
month in San Francisco CA. (She is also, as some of you will already know,
on of our contributing authors in the special World Transport Policy &
Practice number we are dong on carsharing.)

Getting to the Praxitèle site from the middle of Paris is a snap.  All you
up to do is go to the Montparnasse rail station and take a local train to
Saint-Quentin-en-Yvelines, and half hour later upon the alighting the train
and entering the station you will see that a Praxitèle sign is fairly
clearly marked.  Turn your right, go about 20 meters and there you will see
the dedicated Praxitèle parking area and anywhere from six to a dozen
attractive small vehicles (Renault Cleo, 4 doors) parked and in various
stages of being ready to go.  Of course you will probably not have your
Praxitèle smart card already, so it will be necessary for you to continue
about 50 meters down that street and there on the right you'll see the
Praxitèle customer information center.  Once there you will fine someone who
will be pleased to arrange for your visit to the system and a demonstration
run on one of the vehicles. (An advance call, of course, is suggested. (Call
+33 (01) 39 44 92 21 and be prepared to speak French.  Otherwise give us a
call and we'll see if we can serve as an intermediary to organize your well
worthwhile visit.)

That is what Susan and I did. ). We met with and were greatly helped by Dr.
Michel Parent, one of the major forces behind the project since its outset
from  INRIA - Institut National de Recherche en Informatique et en
Automatique ([log in to unmask]) and  Mr. Pierrick Tancray, Directeur
d'Exploitation (Director of Operations), who has been with the project since
September.

Before getting on the street, our hosts gave us a good presentation of the
overall system organization, including its impressive software and support
hardware.  I have to say that the whole thing is quite impressive.  Of
course on one hand one might say  that it's only to be expected  if you
spend a million or two million dollars doing something along these lines.
That said, however it strikes me  that we do have here when it first glance
appears to be a first-class carsharing support system and it is no secret
that developments of the sort, properly done, take time and cost money.

Here briefly are in bullet form some the main highlights and points that we
noted:

* Fifty small, attractive, nicely maintained four-door electric vehicles
* distributed between 9 service points covering the service area
* serving an industrial/residential suburb in a New Town of some 140.000,
grouped into 7 main communes or town centers
* offering 24 hour service
* tracked and polled frequently by a GIS location system
* supported by an office and control center with a first class supporting
hardware and software
* a staff of 4 or so, of whom two occasionally go out to reshuffle vehicles
(manually) in response to signals from the software in an attempt to ensure
universal availability at all 9 depots
* and working largely as advertised.

One of the difficulties they seem to be facing on the supply side is that of
relatively high maintenance costs: the figure all of 40 percent of total
operating cost was mentioned,
* A substantial portion of these cost appears to be a result of the choice
of electric vehicles rather than some more robust technology
* They report very few problems due to accidents or vandalism -- this
despite the fact that the service area is a pretty tough industrial
suburban.   (The Praxitèle team indicated that they thought this was results
all the fact that the vehicles themselves are nicely visible and generally
considered to have a positive social role within the community.)

Thus, what we have here, is a state-of-the-art carsharing operation that in
terms of its size, technical characteristics, and management gives every
appearance of meeting fully the success criteria for such operation.  That
the supply side.  Now a few words on the demand-side.

* Approximately 700 people have purchased Praxitèle cards to this date
* and of these it is said that approximately 400 constitute the basic client
group.
* The 50 vehicles are however presently generating somewhat less than 100
trips per day.  That is the bad news.
* The better news is that as a result of more aggressive management over the
last several months ridership has begun to increase at much sharper rates in
the past.
* The person responsible for the management team (Mr. Tancray) indicated
that he understands that the two key tasks to which he has to give his full
attention and energies are
(a) quality and cost control on in all aspects of the demand-side, and
(b) aggressive marketing to build up those ridership figures.
* On the marketing side, they are currently making an effort to better
identify the system in visual terms, including the vehicles, the stations
and collection points, the central Praxitèle office, and signing around
community.
* There is also a local publicity and information program which is getting
their increasing attention.
* They also indicated that they intended to engage possibly in community
outreach programs in order to make the system better-known and more
intensely used by certain specific target groups.

Final Conclusion:

Certainly one of the most interesting CarShare projects currently underway
anywhere, and if you have to visit not more than a handful this one should
be toward the top of your list.  The project is notable both for its
accomplishments, and for what yet remains to be done.

One wonders about their hardware, software and operations support system.
They have developed this at considerable cost and it gives every appearance
of doing its job notably well.  They also are continuing to work on and
improve it.  Should they be selling it to other start-up ventures across
Europe.  Does that make any sense?  Should we be asking them to do an
article on just that for the forthcoming World Transport Policy & Practice
number?  Strikes me as at least worth a thought.

One final thought that comes to mind is that of how somehow to link all this
knowledge, experience and accomplishment to the several small carsharing
pilot projects that are trying very hard to get on line elsewhere in France.
One would (or at least I would) like to see strong interaction and support,
not least because of the very real difficulties that any independent startup
faces in areas of socio-technical innovation such as this.

That in a nutshell is the Praxitèle status as of early December 1999.  The
notes you have here have not however been as yet run by Susan Shaheen for
her comments and corrections, but we will be putting this before her and
hope that she may have a moment or two to remedy the most outrageous of the
errors or omissions of the above.  Likewise, I intend to share these notes
with the Praxitèle team and if they have any additional comments or
corrections of these will be inserted in the Web site to Susan become
available.



English language summary of some of the technology background taken from
Praxitèle Web site

Praxitèle integrates a large number of technologies to offer the best
quality of service at a reasonable cost. Some of these are :

* The non-contact smart card that offers an easy access to the cars, and
facilitates fare collection.
* The automatic connection with the induction charger. The user is not
bothered about cumbersome and dangerous cables.
* Electric vehicles are now produced in quantities by some car manufacturers
and in particular, by Renault. They are derived from a popular model which
is electrified with Cadmium-Nickel batteries. Later on, more advanced models
designed specially for self-service in city centers, will be used.
* Real-time digital communication between the vehicles and the management
center.
* On-board electronics to control the vehicle, to help the driver and to
calculate the cost of the trip.
* Mathematical models optimize the redistribution of cars among Praxicars,
and the recharge process of electric vehicles.
* Multimedia terminals installed in the Praxiparcs offer a good assistance
to the users such as to call a vehicle when required.
* Long term research is carried out on new types of vehicles that could
incorporate driving aids and even full driving automation.


A Word on How this Note Was Produced

What you see here, warts and all, is an attempt to make use of some of the
latest available voice recognition technology in order to see if I could
quickly share with those of you who might be interested in carsharing the
results of a visit last week to the Praxitèle project in the greater Paris
region.  I chose this particular set of tools in part because I happen to
have a major time problem at this point. But also because I thought it might
be interesting for some of you to see how this reasonably affordable
technology can be put to work in our daily circumstances.  The product I'm
using to do this by the way is the mid range Dragon NaturallySpeaking
package (http://www.dragonsystems.com), and, as you will see, for better or
worse.


Eric Britton

EcoPlan International -- Technology, Economy, Society




%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%