Current parking facilities are extremely inefficient in their consumption of scarce urban and suburban space. A variety of technologies already in use in the transportation industry could make practical much higher density, reliable, automated vehicle storage. Readily available automatic billing systems could also provide powerful incentives for using environmentally friendly vehicles. It is quite possible that these installations could pay for themselves in a reasonable period of time, but their major advantage may be in avoiding the necessity for far more expensive alternatives. The rationales for accepting the mechanical complexities of this type of prefabricated, automated system clearly lie in a) encouraging the use of public transportation systems; b) freeing up other valuable urban real estate for more highly productive commercial development, c) eliminating the need to develop additional rights of way to avoid gridlock and urban congestion and d) providing a practical means to reduce urban air pollution. Unbeknownst to NARPAC when this was written, there is already one European-designed "robotic vault parking system" operating in DC.

(Since this second article was written, NARPAC has been informed that there is another, and more general purpose, automated parking system already in operation in Hoboken, NJ, and built by a firm actually named "Robotic Parking, Inc". Its system is described in a section following this one.)


It would appear to be the heighth of folly to assume that Americans will soon turn away from their love affair with automobiles. Within the metro region, well over 90% of the households in the region own cars, and those that do own cars frequently own two or more. In fact, almost 20% of regional households with cars own three, four, or five. Although these fractions decline somewhat in and near the "inner city", the basic reason seems more closely related to lower household income and size than to any unique urban culture.

By the same token, among all metro area commuters, more than 80% still commute by car, and over 80% of those drive alone to and from work. Half of all commuters work in a different county than where they live, and half of those commute to a different state (or DC). Not to be outdone, some 20% of DC workers also "reverse commute" to the suburbs. A major question for urban and suburban planners, is what to do with all these cars when they are not in use.

It is suggested in several places around this web site that more attention needs to be directed towards the storing of personal vehicles when not in use (particularly downtown) both to encourage the use of public transit, and to avoid the cluttering of limited public rights of way with dormant vehicles. For instance, the subject of parking is hardly mentioned in DC's existing long-range transportation plan, and the subject of pollution reduction is ignored completely. NARPAC believes that trying to design an American urban 'cityscape' which focuses on denying the ownership of private cars is futile.

Most American (barely) adults place a down payment on a car way before they buy into a home, and most Americans live, work, and play in places only accessible by private cars. Surely there are more cars in the US than mothers, or even apple-pie eaters. Car-denial goals are likely to skew urban populations towards those who cannot afford cars, when in fact cities need more resident taxpayers who can afford and do own several cars! The trick is not to limit their ownership, but to limit their inappropriate use and to encourage a switch to environmentally cleaner technologies.

Parking facilities currently consist primarily of open, surface level parking lots, or at best, big (ugly) open garage buildings with several "floors" (above or below grade) of parking spaces accessible to the cars' drivers. Almost all of these lots accommodate all sizes of vehicles, and for that matter, all sizes (height) of drivers. The use of modern technologies is limited to automatic ticket dispensers and gate openers: even the use of automatic billing (like "E-Z Pass") is minimal, and the notion of car ID systems (like bar-coding) is still in the future. Such technologies could easily be used to recognize the car's home garage, for instance, or determine its size, weight, and fuel economy (for variable billing); or match the driver to the car (for security purposes).

Metro's recent Core Capacity Study points up the need to add some 33,000 new parking spaces to the 55,000 already maintained by the Metrorail system to encourage ridership. The vast majority of these spaces are in open surface lots, where the average density is little more than 100 spaces per acre. NARPAC believes that in many locations, this could be raised to almost 2000 spaces per acre by the acceptance of available new technology, and that the required investment could be recouped within five years using proper fee schedules and incentives.

While it may seem academic, treating all "parkable" vehicles as the same size, and requiring (seemingly obviously) that they all be accessible all-around by basketball players, results in very low density use of scarce urban space, and in rapidly disappearing suburban space both in area (footprint) and in volume (which factors in clear height, and floor thickness where applicable). For instance, if all parkable vehicles were packed snugly in one of three "boxes"(like shoe boxes) labeled "small", "large", and "accessible" their footprint would vary from 112 to 210 sqft. In multi-floored garages, their volume becomes key, varying from 672 to 1216 cuft if stacked like cartons in a warehouse or containers on a ship. However, if interchangeably parked with "normal" driver access in multi-floored garages largely devoid of supporting pillars, then their required volume grows to 2560 cuft, not counting the driveway area required to get access to each space. Can technology improve on this, and offer some incentives as well? Obviously. The photograph below, taken on Level 3 of a Bethesda, MD public parking garage shows how little of both the horizontal area and vertical height is actually occupied by the parked vehicles, and suggests how different the size requirements are for a typical compact Volkswagen and the full-sized SUV it faces:

Planning/Design Goals

NARPAC suggests the following long-range goals for planning and planning purposes:

o seek significantly higher density for temporary and long-term vehicle parking;

o adopt technologies currently in practical use by the transportation industry worldwide;

o provide maximum security for the stored vehicles and no shuffling of stored vehicles;

o provide quick "drop-off" (1-2 min) and "pick-up" (2-5 min) for stored vehicles;

o use high energy-efficiency systems wherever power consumption is required;

o use fee/fare systems (e.g., 'smartcards', 'E-Z Pass') consistent with other public transit systems;

o use modular construction to avoid on-site disruption, encourage variable-sized facilities;

o develop automated systems defining owners and significant vehicle characteristics;

o include automated, vehicle/owner-specific fee/fare systems; and

o achieve the same operational reliability as the vehicles being stored.

Suggested Technical Approach

NARPAC has done some preliminary design work, based on the engineering background of its members, and using the following five basic elements:

o automate movement of vehicles from separate "drop-off" to "pick-up" points using pallets;

o store vehicles in generally inaccessible "containers", perhaps matched to vehicle size;

o move pallets by regenerative electrical power (similar to Metrorail components);

o avoid hydraulic systems susceptible to leaks; and

o achieve high reliability through redundancy, component monitoring, and preventative maintenance.

NARPAC Notional Configuration

NARPAC visualizes a standard module of four containers clustered around a central "elevator" which lifts (or lowers below entry level) bar-coded pallets, each with a vehicle secured on it. The driver parks his car on the pallet at the entrance, and retrieves his car from that pallet at the exit. A simplified schematic is shown below:

These pre-fabricated modules could be stacked perhaps 15-20 high, and nested next to one another (and locked together) to form:

o large independent structures;
o the cores of buildings surrounded by apartments or offices;
o "air rights structures" to replace curbside parking above streets or parkways; or
o large subterranean units below government or commercial buildings, public parks, or arenas.

The object is to substantially increase the density of car parking in common systems to which the driving public can become accustomed and hopefully attached, like Metro itself.

A simplified diagrammatic floor plan is shown here involving a square space of 140x140 ft. which equates to roughly half (45% of) an acre. This is the smallest practical size for a typical modern multi-level "square helix" accessible parking garage where a two-way driveway spirals up (or down) with nose-in parking on each side of the ramped sides, and on the outsides of the level ends. A half-acre surface parking lot of this configuration would hold 46 vehicles of any size. Filled with compact cars (Honda Accords?), it amounts to an overall area efficiency of 26%. If this lot is converted to a multi-level drive-in garage nominally 50 ft high and 20 ft deep, it could accommodate 300 vehicles. Again, if filled with compact cars (Toyota Camrys?), it achieves a volumetric efficiency of 14.7%. Even filled with huge SUVs (Ford Excursions?), that volumetric efficiency rises only to 26.6%. The major contributor to this inefficiency is the depth of each floor, requiring 3-ft deep I-beams to span the 50-ft wide unsupported ramps, plus nearly a foot of concrete.

In the notional NARPAC design, vehicles are placed on pallets (nominally 1-ft thick) where they enter and are moved to the nearest elevator which raises or lowers them to an available container. Each cluster of stacked containers has its own elevator, providing redundancy and multiple entrance/exit points. One might imagine a prefabricated building with five levels above a full- height entry plaza and three levels below (to match the 50-ft above ground, 20-ft below ground diagram above). This would hold 536 SUV's, or with eight levels above, five levels below, up to 1350 compact cars. These equate to a volumetric efficiency 47.5% and 66.1% respectively. One each of these storage buildings could thus hold almost 1900 vehicles per acre compared to 102 for a surface lot, or 636 for a rectangular helix parking garage. As a plausible application of this technology, the chart below shows that a one-acre, 5-level (total) containerized garage could hold all the vehicles curbside parked for nine full city blocks (400x600ft).

NARPAC Notional Components

The following components, only partially worked out, are described, with the source of the needed technologies:


Drive-on aluminum "stackable" pallets would be used with groves and chocks for vehicle wheels, and hard points for lifting, pushing, supporting and stacking pallets. Nominally one-foot thick, pallets would have almost no mechanical parts.

Pallets would have permanent bar-codes, so the system can identify the storage and return destinations, and note them on the "car-check card". (technology: common tow-truck beds; railroad boxcar, postal package shipping bar-coding systems). A few special "patrol pallets" would be suitably equipped for inspectors to "ride" through the automated system doing trouble- shooting, routine maintenance, and/or repair work.

Pallet Roller Track Systems

Standard lengths of pallet roller frames would be used, using two rows of rubber-tired automotive wheels powered by electric motors to move the pallets along, riding in groves on the underside of the pallets. Pallets would trundle forward above their pallet tracks, or sideways to change from one track to another. Some kind of braking system, hopefully regenerative, would also be required for each wheel. (technologies: auto/truck chassis, wheels, and brakes, Metrorail electric motors, overnight package delivery systems)

Pallet Stackers

Slightly modified pallets would be used to permit stacking and moving of 5-8 empty pallets, since one is needed per stored vehicle and each must be moved back during off-peak times from "output" gates to "intake" gates (not an insignificant problem). Procedures would be developed to collect/stack pallets of departing cars and move/restack them for entering cars during off-peak times. Stacks would be stored in containers closest to the "ground floor plaza" to permit rapid recovery and return. (technologies: as above)

Storage Containers with Vertical Cog-rails

Steel or aluminum containers of minimum realistic size and clearance would be used, half would be end-loading and half side-loading, each capable of "receiving" and locking a pallet, pushed over to it by the vertical cog-elevators (below). Each container would have reinforced corner posts for stacking loaded containers up to fifteen high, and include some bar-code reader identifying the pallet it is storing, and some distress (motion, smoke, heat) detection systems. Vertical edges of each container would be reinforced for attaching vertical cog-rail guide-ways for cog-elevators (below) (technologies: container ships, cog railways, household security systems);

These stacks of containers would appear very similar to those in everyday use on and below deck in container ships. For aesthetic purposes, some additional exterior surface could be applied for specific installations. There is no requirement for external ventilation since no internal combustion engines are involved inside the structure. The construction of the vehicle containers differs slightly from shipping containers in that no load-bearing floor would be required . In this regard, the car containers would be more like metal file cabinets where the load in born by rollers near the side walls (i.e., the pallet equates to the file drawer). A crude 3-D schematic of the system is shown to the left.


The key mechanical element in this notional system is the cog-elevator proposed to lift (or lower) each loaded pallet to its designated receiving container, by "crawling" up corner cog-rails using gripping cog wheels (positively synchronized) at each corner and driven by 75-horsepower Metrorail regenerative electric motors. Each elevator would include lengthwise pallet roller frames to receive and discharge pallets. It must also include raisable side-wise pallet rollers for moving pallets for storage to either side. Each must also include mechanisms for pushing pallets to, and retrieving them from, positive detents in each container (and for sensing if a pallet, by mistake, is already occupying that container). Aircraft electric rotary screw shaft flap extenders might be used. Metrorail motors maximize initial torque, consistent with the needs of these elevators.

In tall structures, two or three elevators might be included in each shaft to reduce storing/retrieving times, and hopefully to provide redundancy for the system's most complex component. Components would be ruggedized, and 'instrumented' for predicting potential failures. Some might be able to visualize these elevators as having evolved "genetically" from contemporary 4-wheel drive tow trucks. (technologies: metrorail power systems, vehicle drive-shafts, recovery-vehicle details, commercial aircraft auxiliary power systems).

Power Towers

One reason for using electrically powered systems, besides their own environmental cleanliness, is to be able to convert kinetic energy from the descending loaded cog-elevators back into potential energy. This would be done by using the re-generated electricity to lift some very heavy weight in a special elevator shaft. The potential energy of that raised weight would generate electricity when later allowed to descend. The "packing box" in which the vehicle is moved about on its pallet still has a "density" of only about 5 lbs/cuft (a 3000-pound car in a 600 cuft "box"). Solid steel, on the other hand, has a normal density of about 500lbs/cuft (100 times higher). Low-grade recycled steel from crushed cars could be used to provide the heavy weights. Hence, with perfect power conversion efficiency, a (stronger) cog-elevator could easily raise a weight sufficient to then provide power to raise 200 vehicles (half as high on average). With only 75% efficiency, it could still lift 150 cars, or, say, 100 cars, plus providing power to move the pallets horizontally in and out of the parking structure. (technologies: certain heavy industries, including mines, and Metrorail cars using electric regeneration)

Computer-based Management System

The computer management system would have two basic functions: first, of course, to manage the movement of the loaded pallets to and from their storage containers on demand. The container could easily be selected based on short- or long-term parking, vehicle weight, etc. That card would subsequently be inserted by the owner into the pick-up slot at a suitable delivery station. This aspect of the system is based on tracking and moving the pallets, independent of the vehicles on them. (Technologies: automated warehousing, package mailing, delivery systems)

Second, the computer system would do the billing and issue to the driver a "car-check card", in response to a credit card or "metrocard'. It could determine individual storage fee rates depending on many vehicle- and owner-specific variables. Note here that the fee structure is based on the characteristics of the vehicle and the address of its owner, not the pallet. (technologies: supermarket check-out systems, Metro "smart cards", E-Z Pass technologies). These variables could well include all the following:

o How long has the vehicle been stored?

o Is this a fuel-efficient (or hybrid, say) small car deserving favorable parking rates?

o Is this car being returned to its long-term home-garage? (Another favorable long-term rate)

o Is this car being parked at its normal daytime place of employment?

o Is this parking facility at a normal transfer-to-public transit location?

o Is this parking lot in the city, and is this car registered in the suburbs?

o Is this cost to be shared by several arriving persons' metrocards (i.e., carpooling)?

o Is this charge to be billed to a credit card or deducted from an existing deposit?

One could easily visualize a home-parking fee downtown for a small hybrid import car for $2 per day, or a downtown parking fee for a gas-guzzling SUV visiting from the suburbs for $40 per day. In between, a "family-sized" sedan might be stored at a Metro station just outside DC for $9 per day, split between its three occupants, or for $12 per day for a sole occupant. Parking fees can become an important source of incentives for intelligent automobile usage, as well as providing substantial revenues. A notional fee schedule is shown below:

For whatever its worth, London has recently introduced a "congestion charge" of 5 pounds per weekday (almost $8) for all but exempt or designated "green cars". In a rather extraordinary and seemingly awkward system, drivers are encouraged to pay in advance for trips into the Central City. All entering/exiting roads within that area are under continuous surveillance by cameras linked to "automatic number plate recognition computers" that check to see if that car has pre-paid for its entry. (DC now uses this technology to catch speeders and red-light runners.) Stiff fines are imposed on the car's owner if the fee is not paid before 10:00PM that day. If the driver has bought entry but not used it, there is another time-consuming process to get back a portion of the fee paid. This system is being operated by a contractor under a six-year deal for an undisclosed sum! It has one advantage in that it applies to any vehicles that enter the prescribed zone even if they don't park

System Costs and Benefits

Clearly, these NARPAC-proposed prefabricated, computer-driven mechanical systems will cost substantially more than macadam paving, poured concrete buildings, or surveillance cameras. Notional cost estimates for the various components listed above are included on the table below. Summing the parts needed for the 140x140 half-acre structure described earlier and dividing by the number of cars accommodated, indicates that a "metropark" building for SUVs might cost $8000 in parts per space provided. Assembly and erection costs might double that. Costs might be slightly lower based on the smaller volume required for smaller cars, but the installed cost per space, even in larger, taller structures is unlikely to fall below $15,000. Amortized over five years, and assuming each space is used 200 days per year, a base fee of $15-$20 per day would be required, barring subsidies as inducements to use public transit. Amortization over 10 years could reduce these costs to $8-$10 per day. Investment costs for high-density parking facilities could easily run $30M-$40M per acre.

It is a also possible to make a rough estimate of the potential revenue returns per acre for the two 50-ft high, 20-ft deep configurations previously used as an example. For two automated half-acre garages, one for small vehicles and the other for large ones, the chart below indicates what the annual returns and uses might be for a downtown garage. If the garages both sustain a 60% weekly space occupancy rate (which might result from a 75% weekday fill, plus a 20% weekend fill), if the average SUV fee is $20, and the average compact fee, $12, then the garage would collect $16,000 per day, or $5.9 million per year. $1.5M might be allocated for city taxes, $840K for maintenance (= 20 man-years?), and $3.6M for depreciation. These are very substantial funds, albeit very rough estimates!

Given the public needs, however, there is no burning requirement that these parking facilities pay their own way. NARPAC cannot quantify the opportunity costs resulting from lower congestion, less pollution, and less curbside parking on main roads. However, the savings in valuable urban ground space, particularly around metrorail and metrobus transfer stations could be significant. As mentioned earlier these metropark sites could raise parking lot densities from 100 spaces per acre to at least 2000 spaces per acre, while staying well below allowable building heights. As an extreme, then, one high density acre of parking could replace 15-20 acres of surface parking (There are 25 wasteful acres of such surface parking around the US capitol itself .

According to NARPAC's "net productivity" calculations, the average commercially zoned acre in DC produces net revenues of almost $700K annually, while the higher density developments can net about $2 million in various tax receipts. While the estimates provided here (by proponents) are doubtless optimistic, it suggests that one acre of metropark facilities might cost $42M to build and the 14 acres not used for surface parking might net the city $14M annually in revenues. For the city to get its investment back in three years may be somewhat delusional. However, the chances of getting it back in five years seem excellent, particularly if reasonable vehicle-dependent fees are charged.

Alternatively, one might consider the probable costs of developing additional roads within the city. DC property assessments add up to roughly $48 billion for roughly 12,000 acres of taxable land, or $4 million per developed acre. A linear mile of 10-ft wide right of way amounts to 1.2 acres, and might have some 200 parked cars along its length. Buying up and building a one-lane mile of in-town street could easily come to $10M, even without the costs of reconstituting impinged buildings or utilities. This compares to buying and building a quarter-acre automated car park for those 200 cars that might run $10M tops. Creating parking facilities seems far cheaper than creating new urban rights of way. They would be particularly attractive when combined with the emerging need for dedicated-lane "busways".


Current parking facilities are extremely inefficient in their consumption of scarce urban and suburban space. A variety of technologies already in use in the transportation industry could make practical much higher density, reliable, automated vehicle storage. Readily available automatic billing systems could also provide powerful incentives for using environmentally friendly vehicles. It is quite possible that these installations could pay for themselves in a reasonable period of time, but their major advantage may be in avoiding the necessity for far more expensive alternatives. The rationales for accepting the mechanical complexities of this type of prefabricated, automated system clearly lie in a) encouraging the use of public transportation systems; b) freeing up other valuable urban real estate for more highly productive commercial development, c) eliminating the need to develop additional rights of way to avoid gridlock and urban congestion and d) providing a practical means to reduce urban air pollution.

return to the top of the page IS REALITY STRANGER THAN FANTASY?

The Only Robotic Parking Garage in the US is Already Operational in DC!

Shortly after the preceding presentation on high-density parking was published on our web site, several knowledgeable readers suggested that we check out what appeared to them to be a similar system already installed in a remodeled downtown DC residential building. Sure enough, NARPAC's creative notional 'invention' appears to have been already perfected and in large scale use to solve serious urban space limitations in European cities designed and built hundreds of years before the advent of the car.

In fact, the first use of this type of robotic parking design in the US was occasioned by the DC requirement to provide parking spaces in all new and renovated residential buildings. The Summit Grand Parc is the latest reincarnation of a 'gracious' Italian Renaissance style building on the Southwest corner of McPherson Square (15th and I Sts). It was built in 1912 for Washington's University Club with marble staircases, main floor lounges with 20-ft ceilings, etc. It was bought in 1936 by the United Mine Workers, and a penthouse suite was added that became the permanent residence of the well-known union boss, John L. Lewis.

The original 5-story building was placed on DC's Historic Preservation list in 1999. Summit Properties has now wrapped a new "residential tower" around the off-street sides of that historic corner building, now converted to commercial/retail use, to provide 104 luxury apartments. Within the 60x106 ft extension on I Street, Summit was obliged to include 74 underground parking spaces: a virtually impossible task without adopting an automated high-density approach, which is now featured as a "robotic vault parking system". Installation was done by the "SpaceSaver Parking Company", a subsidiary of the Mid-American Elevator Company, using the basic European designs Wohr-Stopa Corporation in Stuttgart, Germany which can be tailored to individual space (volume) demands. This constrained design achieves a rather remarkable volumetric storage density of one vehicle per 2750 cuft, compared to 5100 cuft for a large, optimally configured, drive-in parking garage (see prior section) , or as much as 12,000 cuft per vehicle for this inefficient volume.

Design Details

The below-ground "vault" consists of two parallel rows of eight "racks", either four or five shelves high, into which pallets can be slid nose/tail first from a traveling crane that moves between the rows, depositing or retrieving a pallet from the computer-designated stall. The installation is nothing short of a mechanical marvel of steel parts, miles of greased belt/chains, adjustable Teflon rollers, and many electric motors. It handles its Mercedes, BMWs, Porsches, and Audis with dignified, measured, humming and clicking precision. The only thing missing is the background music. In this installation, there is no attendant on duty! Furthermore, in a nice touch, the elevators of the residential tower are equipped with card-insert devices which allow the resident to save time by calling for his/her car on the way down.

In this particularly cramped installation, cars enter through a remote-control door to a small alley with two identical diagonal bays also enclosed behind remote-control roll-top doors. The computer system opens the door to the appropriate bay, the driver drives the car onto a flush pallet on a turntable. Extensive, automated exit procedures are followed ("take your bags", "lock your car", "watch your step", "grab your pet", etc.); a battery of laser beams measure the car's dimensions to assure its proper positioning and a suitable storage destination; separate sensors and controls determine that the car's occupants have left the bay; lower the door, rotate the car to align with the down elevator; and lower the car one level to an upper rack space where the rail- traveling crane draws the pallet aboard and moves it to the nearest available, correctly sized stall. As a thoughtful detail, when retrieved, the turntable rotates the car to face outward. In one telling aside, this current European design is not large enough to handle the largest US cars and SUVs.

Some of the graphics in the SpaceSaver Parking Company brochure are shown to the left: a simple schematic of a nose-in storage system using an overhead traveling crane; a large rack of nose-in pallets is shown from a typical European installation; a side-loaded tall stack is shown from the elevator shaft; and a novel vertical showcase for full-sized Saab cars is used in an urban European showroom. It is clear in the area of parking that the more crowded European cities are well ahead of their more sprawl-tolerant American counterparts.

One complicating innovation of interest to NARPAC is the handling of the empty pallets (an issue poorly resolved in its own notional design). In this SpaceSaver design when a loaded pallet is delivered to its stall, an empty pallet is first removed from that stall. This seems to require a double-level pallet track in the rack, and the movement of the empty pallet from one track to the other. In the process of shifting that empty pallet, it is also tipped up to drain into rack-mounted gutters any water that may have collected in the pallet's wheel channels from a snow/rain-soaked car. There is a very weak analogy to the wet trays at the entrance to most fast food cafeteria lines.

Despite the very complex system, it was installed for roughly $20,000 per space, and is rented to residents for $300 per month. It was not intended to be a money-maker, but rather a unique solution without which the building could not have met the relatively modest zoning requirements of three-quarters of a car per resident household (see NARPAC's section on regional and city car ownership). NARPAC imagined an $8000 cost per space in a large, optimally configured space but surely would not have accepted a fixed price contract to build it! More detailed comparisons are provided below for the technically minded reader.

The amateur NARPAC photograph below shows the ongoing storage of a Saab convertible that had just been driven into the upstairs garage and left by its owner in the hands of the fully- automated system. The brown corners of the photo result from cropping, not the structure of the building. The pallet with the car on it is just disappearing into its storage stall to the left. It has been computer-directed to the closest stall of the smallest size compatible with this car (in this case, still two shelves above the floor level). The empty pallet has already been removed from the rack and rests below the rollers on the floor-mounted, blue-painted traveling crane. The lighting was on simply because human observers had entered the otherwise unpopulated space. The pallet could equally well have been off-loaded directly to the lower right side, (one level below the photographer). Note that the total building width here is just sixty feet, including the foundations. These mechanical structures are specifically designed and tailored from standard components for each installation.

Comparisons to the NARPAC Preliminary Design

There is some dubious arrogance involved in comparing the Woehr/SpaceSaver (WSS) reality with the preliminary NARPAC fantasy. Nevertheless, it provides a useful academic exercise which could prove useful in future real designs:

The two concepts derive from different backgrounds:

The WSS clearly appears to have evolved from an extensive background in the development of automated warehouse devices. This is still evident in their other product lines for stacking and retrieving anything from board lumber to furniture suites. Early designs for such robotic "stock rooms" date back well over thirty years and clearly point towards pallet-on-racks designs, using traveling cranes with chain-driven hoists (as on fork lifts) and available electric power. These designs are also most frequently applied as equipment within existing buildings (as the example above) and still find extensive applications for relatively small , widely varying storage demands as perhaps in Home Depot, for instance).

The NARPAC notional design started specifically for large numbers of cars, in free-standing structures, with high periodic daily demands (as for commuters), in which a high, if not too high, premium was placed on using direct automotive technologies and parts. The maximum possible volumetric efficiency was sought to minimize "urban sprawl". We also wished to emphasize the need for energy efficiency in the garage system, and to proffer incentives for fuel efficiency in the vehicles being parked.

Several basic elements are essentially identical:

o the use of pallets to provide a standard interface between garage and vehicle;
o the use of minimally sized, selective, limited access storage stalls;
o the use of relatively clean electric power;
o the use of "dumb" pallets and storage bins, concentrating the complexity in a few moving components (elevators and/or cranes),
o card-entry, computer-dependent storage,

But several are significantly different in the NARPAC approach. The following codes apply:

"o" appear capable of incorporation in either design;
"+" seem like NARPAC design-specific advantages for cost, speed or compactness;
"--" seem like possible NARPAC higher risk or shortcomings

o using 'containers' rather than open racks to provide structural building blocks for stand-alone buildings, many stories high/deep;
+ proposing pallet storage on all four sides of a single elevator shaft for max density;....
+ .....thus requiring more vertical elevators, but no traveling cranes;
-- using "cog-elevators" crawling up fixed cog-rails, rather than very extensive chain drives.....
+ thus allowing more than one elevator per shaft for faster storage/retrieval;
-- using rubber-tired roller wheels rather than more compact plastic rollers.....
+ but proving powered roller tracks to move pallets horizontally;
o building in vapor detectors against vehicle gas leaks;
o built-in preventative maintenance and maintenance pallets;
o using regenerative electric power to conserve energy;
o assuming prior sorting of cars to fit pallets, not using laser measuring devices;
o relating fare systems to vehicle home address, size, type;

-- requiring extensive pallet-stacking by not swapping full/empty pallets at each bin;
+ but providing for on/off-loading of pallets away from elevators for faster storage/retrieval;
o using bar-coding on pallets to identify vehicle destinations;
o assuming (though not specifying) human attendants always on duty for safety, maintenance, etc.
o using container floors and walls to keep each stall from endangering/dirtying/wetting others.

The most interesting to NARPAC of these differences is the space-saving, but time-consuming WSS method of storing an empty pallet in each empty stall. It would appear to be an obvious solution were it not for three considerations: 1) moving the empty pallet on and off the elevator/crane takes as much time as moving the loaded pallet, thus doubling that segment of time; 2) the complexity of shifting that pallet from its "full track" to its "empty track" seems to be excessive in the current WSS design (and the pallet tipping kinda silly), and 3) it is perhaps fatally compromised if the loading and unloading is done at some entry and exit plazas apart from the elevators/cranes. The NARPAC design envisions collecting and stacking pallets at exit plazas where commuters retrieve their cars, and then moving those stacks back to the entry plazas where they would be filled the next morning. Much like clean cafeteria trays find their way back to the beginning of the line for reuse. The trouble is that there can be a very significant build up of pallet stacks that need to be stored somewhere: NARPAC stores them in empty stalls and essentially shuffled them around like loaded pallets (or maintenance pallets), returning them to the entry plaza as needed.
The riskiest part of the NARPAC design may be the cog elevator which eliminates virtually all of the requirements for well-aligned and lubricated chain drives all over the installation. The question seems very simple: can the vertical cog rails be so stiffly and permanently aligned that they never splay out allowing the drive wheels to lose their grip. NARPAC tried to solve this problem by using "grabbers" to hang on to the back of each rail, essentially assuring contact between gear and rail teeth. We do not know for sure if this is high-reliability engineering.


In September of 2003, NARPAC received a missive from the president of Robotic Parking, Inc. of Pinellas Park, Florida, asserting that America's first robotic parking garage was not in fact the work of the SpaceSaver Parking Co. (described in the previous section), but the creation of Robotic Parking (RP). It has been fully operational since May, 2002 on Garden Street in Hoboken, NJ, a city better known as the birthplace of Frank Sinatra. 324 cars can be parked on seven levels in a building with only a 100'x100' footprint.

The company uses many of the same arguments that NARPAC has presented in the opening section of this chapter, but makes a far more professional presentation on their web site at They cite 10 major advantages of such parking facilities, ranging from more efficient land use and mechanical operation; reliability of the system; and safety and security for the both vehicle and its occupants; to all-electric operations, with no hydraulics that leak; redundancy in off-the shelf components; and a fully American-produced system (though it has its roots in Western Europe, where such parking systems are now commonplace). A schematic for a full-blown installation offering for over 110 cars per floor is shown below:

The layout is rectangular with a central "aisle" serving spaces two-deep on each side. RP prides themselves on separating the functions of rotating the cars from front-in to front-out, traversing the aisle horizontally on each level, and providing vertical lift shafts. As in NARPAC's vision and the SpaceSaver design, the car is left on a pallet by its driver, and stays on it until retrieved. Instead of NARPAC's somewhat awkward "plus-sign" layout of four bins served by a cental shaft, this design achieves virtually the same design area density (near 80%) by storing its vehicles in racks two-deep. This requires sometimes moving the pallet nearer to the aisle to get to the deeper pallet, and leads to the use of several traverse travelers on each level, which can remove the blocking car while another traveler retrieves the wanted one. In a garage of this size and layout, more than a dozen palleted vehicles may be in motion simultaneously.

System Operation

The left picture below shows the rather elaborate "foyer" of the Hoboken parking garage where the driver leaves his car, follows a detail list of safety instructions (as in the Summit Grand Parc), exits the room and closes a garage-like door before operations can begin. This is intended to be a fully automated operation with no attendant present. The right hand picture shows the vehicle being rotated on its entry level carrier/traverser.

The vertical elevator design is shown on the left side below, and a typical traverse carrier on the right side. It is of some interest that RP estimates that storing and retrieving one vehicle will consume about one kilowatt hour of electricity which costs about ten cents. They have not tried to incorporate regenerative motors to lift deadweights to conserve energy, however there is no inherent reason why they couldn't if such a feature was desirable either to save energy or to assure operation during blackouts.

The RP system is confronted with one problem (and solved it) concerning the handling of empty pallets. Unlike the SpaceSaver design, which leaves an empty pallet wherever it removes a loaded one (and vice versa), RP pallets are collected, stacked, and stored like cafeteria plates under the "foyer", and dealt off from the top as each new vehicle arrives. Furthermore, the design is flexible enough so that very narrow garages can be built which store cars only in front and behind a single shaft: such an installation might be squeezed into the empty space above alleys between existing tall buildings. Conversely, if the space is wider and shallower, the pallets can be offloaded left and right of the shaft, perhaps to fill wasted courtyard space behind city buildings.


System Costs

RP hopes to construct robotic parking facilities large or small in many of the most traffic- congested cities in the US. Hence, their published cost estimates have to be substantially better than those on the back of a NARPAC envelope. Nevertheless, they make NARPAC's guesses look pretty good. The table below shows the projected investment and operating costs, including land purchase and debt service (!).

Please note that the "typical" baseline parking lot is assumed to be several (4?) levels high and that the rather modest cost differentials would be increased if the baseline were open surface parking as is still so common around DC including Metro lots, and 45 acres of prime national real estate near the Capitol itself. It is also important to realize that in many cases in DC, the "land" is already owned and the need is to increase capacity. Finally, there is no calculation offered in this RP example concerning revenue productivity of the land. When the name of the game includes the generation of city revenues (particularly from commuters), then the arguments for multi-level, high-density, automated parking become even stronger.

RP Design vis-a-vis NARPAC Vision

There are a few innovations in the NARPAC preliminary design which differ from those of RP, but which could, if deemed worthwhile, be added to the current RP system. These include the following:

o RP uses a steel frame structure assembled on the site, and quite cleverly suggests any kind of exterior design appropriate for its location (see photo below). NARPAC was moot on the point of exterior design, but used pre-assembled "containers" stacked on site to lower cost and limit urban disturbance;

o the RP design was specifically designed to take the largest passenger vehicles made today, resulting in a single standard pallet (and hence footprint), and overhead clearance. NARPAC saw a useful possibility in designing at least two different sized garages, one for larger cars and SUV's, the other for economy-sized cars. We placed greater stress on higher volumetric density design for DC because of its rigid building height limitations as well as its scare available land. We also hoped variable parking rates could be used to generate incentives for driving environmentally acceptable vehicles into urban areas;

o There was no particular effort in the RP system design to encourage transfer to public transportation at the city's outskirts. NARPAC believes that these robotic parking spaces may be used more by commuters than by residents, and that wherever possible, the parking garage design should entice drivers to transfer onto Metrobus or rail. Again, RP could add such features (combined "smartcards" for payment, for instance, or increasing fees for out-of-state parkers) if city planners so desired.


In general, this Robotic Parking design and early installations demonstrate beyond question that such modern technology can be used to help in the all-important union between private and public transportation. It can be an important component of smart urban/suburban growth. NARPAC urges DC metro area transportation officials, in conjunction with Federal agencies as appropriate, to begin to incorporate automated parking systems in the near-term city development plans.

return to the top of the
Did you find this of interest?
Please give us your FEEDBACK

This page was updated on Dec 5, 2003



© copyright 2007 NARPAC, Inc. All rights reserved