1. Questions

The city of Zurich (Switzerland) pursues an active parking policy which is aimed at promoting living with a low number of cars (Stadt Zürich 2022c). Changes in parking space availability can influence many parameters in the transportation system such as mode choice, route choice and car ownership rate (Christiansen et al. 2017). This paper gives insights into the parking occupancy and search dynamics in the area between Bürkliplatz and Münsterhof in central Zurich where around 300 on-street parking spots are accessible to the public. The location is close to the historical center and Bahnhofstrasse shopping district and therefore highly attractive for parking. Fig. 1 shows the perimeter (yellow polygon) in central Zurich and parking spaces (red) as a heat map. Parking is charged between 9:00 a.m. and 8:00 p.m., the maximum parking duration is generally two hours with the exception of a short street section with a 30 minute limit. The fee increases with time: 1 Swiss Franc for up to 30 minutes, 3 Swiss Francs for one hour and 7.50 Swiss Francs for two hours.

Figure 1.Perimeter (yellow polygon) and heat map of on-street parking spots (red) in central Zurich, heat driven by number of on-street parking spots per area

Source: Including data from Stadt Zürich (2022b), Stadt Zürich (2022a) and background map from Kanton Zürich (2022)

The parking situation was tracked during a sunny spring morning from 6:30 a.m. to 11:30 a.m. when the farmer’s market takes place on Bürkliplatz. The following questions are addressed:

• How does the parking occupancy develop on a market morning?

• How does the parking behaviour and its perception by the drivers change on a market morning?

• What influence does the farmer’s market have on parking behaviour in the area between Bürkliplatz and Münsterhof?

2. Methods

Two methods were used to track parking search and occupancy. To record the parking occupancy, all parking spaces within the perimeter were inventoried and the perimeter was divided into three subareas, as shown in Fig. 2. The occupancy of individual parking spots was checked every 15 minutes by one person per subarea by noting the parked vehicle’s license plate and, when present, if the vehicle had a special parking permit for market staff. The number of times a vehicle is subsequently recorded allows estimating its parking duration. In total, 871 different license plates were noted.

Figure 2.Parking spot inventory and perimeter subareas

Source: Including data from Stadt Zürich (2022b)

To study the parking search traffic in the perimeter, team members tracked vehicles driving into the perimeter by bike and asked drivers (after they parked and left their vehicle) about their destination, estimated parking search duration and estimated total parking duration. As soon as the car enters the perimeter, they started to record the exact route, distance to parking spot and parking search duration with a GPS application, as well as the vehicle license plate and the parking space number. From 6:30 a.m. to 11:30 a.m., 82 vehicles were tracked by bike and 66 vehicle drivers were interviewed.

At 6:30 a.m., 36 % of the total 296 available parking spaces in the perimeter were occupied. Occupancy increased progressively until 9:30 a.m. when it reached its peak with almost all parking spaces being used. At the end of the survey, 82 % of the parking spaces were occupied. The occupancy of the parking spaces in the individual streets is shown as a matrix in Table 1. In the streets around the Bürkliplatz farmer’s market, three quarters of the parking spots were occupied, on average, until 10:00 a.m. In the same period, only half of the spots were occupied, on average, on the remaining streets, which were not in the immediate vicinity of the market. The lowest occupancy rate has been observed in the subarea where the allowed parking time is limited to 30 minutes, with an average occupancy of 50 %.

Table 1.Parking spot occupancy per street [%] during study period: red street area names are close to the market

Parking search duration increases as parking occupancy increases. The search duration increases until the longest search duration (752 s) for an individual vehicle is recorded at 10:25 a.m. After 10:30 a.m., the search duration decreases slightly until 11:30 a.m. The average search duration was 91 seconds. Other recorded values such as time lost due to parking search, the distance difference between optimal parking spot (directly at stated destination) and final parking spot or unnecessary distance driven behave similarly.

3. Findings

The study shows that parking occupancy increases during the morning (see Fig. 3). Parking spots in the vicinity of the market fill up significantly earlier than those further away from the market. The parking spaces further away from the market function as a kind of buffer as they are used more when occupancy in the area close to the market is high (80 % or more). At 10:00 a.m., only five of the 296 available parking spots are empty. After that, the occupancy decreases slightly, which could be caused by the closing of the Bürkliplatz farmer’s market at 11:00 a.m. Up to an occupancy of about 60 %, the parking search time remains at a constant low level. At peak occupancy, drivers sometimes stop the parking search and leave the perimeter. During the market’s opening hours (6:00 a.m. to 11:00 a.m.), 20 % of all parking spaces in the perimeter are permanently occupied by market staff, in some cases also in the area close to the market. This leaves only 80% of all parking spaces available for other drivers.

Figure 3.Occupancy, search duration and search strategies over time

Drivers often overestimate their own parking search time (see Table 2). The estimated values (expressed by drivers in minutes) usually deviate from the measured value (measured in seconds) by one or up to two minutes. The estimates of parking duration are more scattered. Likewise, most respondents overestimate their expected parking duration even as they are trying to keep their stay short. The deviation from the measured parking duration is usually not higher than one hour. People who underestimate their actual parking duration are in most cases short-term parkers. Some people parked longer than our data collection period (6:30 a.m. to 11:30 a.m.). Without these, the difference between estimated and measured parking duration is lower (-7 min), but it stays negative.

Two parking search strategies of market visitors were observed: Risk-Strategy and Search-Strategy. Applying the Risk-Strategy means waiting at the perimeter entrance until a spot is free or parking directly on a vacant parking place at the Bürkliplatz farmer’s market. Using the Search-Strategy, drivers continue north inside the perimeter and search for a parking place. The Risk-Strategy is the preferred parking search strategy among market visitors before 9:00 a.m. After that, the Risk-Strategy is chosen by only two of the six arriving market visitors. When comparing the two parking strategies, the Risk-Strategy proved to be clearly more efficient, but it is only promising if there is sufficient free capacity in the area around Bürkliplatz farmer’s market. Fig. 3 illustrates the discussed findings.

Acknowledgements

We express our appreciation to Christopher Tchervenkov and Kay W. Axhausen for their advice and sharing their expertise. In addition, we are indebted to three anonymous reviewers for helpful comments on previous versions of the manuscript. All remaining errors are our own.