Occupancy uses reconstructed Mon-Fri 09:00-17:59 hours; event counts require the whole log week. Relative change is hidden for partial coverage or a zero baseline.
Demand and capacity
by hour
Spots taken · Mon-Fri average of 6
Times /park sent · all days · red is turned away · tallest 36
00
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Two different measurements sharing only the hour axis: the top row is reconstructed occupancy sampled on workdays, the bottom is a raw count of commands on every day. They have no common denominator and one does not divide into the other. A missing top bar is an hour never sampled on a workday.
Parking session duration distribution
n=42·4 partial excluded
5
1
3
4
29
<2h
2-4h
4-6h
6-8h
8h+
Observed transition spans only; timestamps are sampler observations, not exact arrival or departure times. First-snapshot, ongoing and partial sessions are excluded.
Availability probability by hour
observed hours
0006121823
Share of observed hourly samples with at least one spot free · hover for sample counts · missing means not observed
When the lot is busy
spots taken
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Mo
Tu
We
Th
Fr
Sa
Su
1
0
1
1
1
0
0
4
3
4
4
3
0
0
4
4
4
4
3
0
0
4
4
4
4
3
0
0
4
4
4
4
3
0
0
4
4
4
4
3
0
0
4
4
4
4
3
0
0
4
5
4
4
3
0
0
4
5
4
4
3
0
0
3
3
5
4
3
0
1
1
3
2
3
2
0
1
0
2
0
0
0
0
1
1
2
0
0
0
0
1
1
0
0
0
0
0
1
0
0
0
0
0
0
1
0
0
0
0
0
0
1
0
6· blank = no data
Day by day
busiest moment
Mo
Tu
We
Th
Fr
Sa
Su
3 Aug to now · blank = no data
Who parks here
days seen
MI2709
arrives ~09:08 · stays ~9.4h
10
NM694
arrives ~08:57 · stays ~9.9h
9
OD1363
arrives ~13:14 · stays ~5.3h
8
LF443
arrives ~08:38 · stays ~8.7h
3
NV3036
arrives ~08:30 · stays ~9.4h
3
08:0024:00
Usual days
days here
MI2709
NM694
OD1363
EX4482
LF443
Mo
Tu
We
Th
Fr
Sa
Su
2
2
2
2
2
0
0
2
2
1
2
2
0
0
2
2
2
0
1
0
1
0
1
1
0
1
0
0
1
0
0
1
1
0
0
of
2
2
2
2
2
1
2
Days the car was in the lot, out of the weekdays this log spans · a blank cell is a weekday the plate was never here on, a dash is a weekday the log has not reached yet.
Recent swaps
MA1547 took LF443
21 Aug 18:40
EX4482 took HL1817
21 Aug 11:20
LF443 took EX4482
21 Aug 08:19
EX4482 took EY5071
19 Aug 10:27
NV3036 took MA1547
19 Aug 07:52
MA1547 took LF443
17 Aug 16:04
EY5071 took MU1825
17 Aug 07:42
LF443 took EX4482
13 Aug 08:37
HL1817 took MA1547
13 Aug 08:14
MA1547 took GM105
12 Aug 15:48
MU1825 took NV3036
12 Aug 08:45
GM105 took HL1817
11 Aug 10:36
EX4482 took LF443
11 Aug 09:21
HL1817 took EY5071
11 Aug 08:10
NV3036 took HL1817
11 Aug 08:09
EY5071 took MU1825
10 Aug 07:47
Who takes from whom
times
MA1547→LF443
2
LF443→EX4482
2
EY5071→MU1825
2
EX4482→HL1817
1
EX4482→EY5071
1
NV3036→MA1547
1
HL1817→MA1547
1
MA1547→GM105
1
Swap balance
lost · took
HL1817
LF443
EX4482
MA1547
EY5071
MU1825
NV3036
GM105
How long a spot stays held
35 holds so far, and none of them got a late re-request.
0%
Needed longer
-
Extended holds
0
Given back
Needed longer counts a re-request late in a hold's window - strong evidence the hold was too short, but the log cannot tell an Extend tap apart from a Re-park tap or a plain typed /park landing late by chance. Extended holds is a range over those same holds, not a measurement: the low end is when someone last asked for the spot, the high end runs to expiry. A hold also gets recorded when the car is already parked, and the log cannot tell that apart from holding an empty spot - so the high end is generous.
When there is no spot
Every one of the 16 park requests found a spot.
0
Turned away
7%
Full, 9-17
0
Errors
Outcomes by week
tallest 31
3/8
10/8
17/8
got a spotheld onlyno spoterrorsother
Raw counts, not shares · the faded column is the week still running · "held only" is a /park for a car already in the list, which never asked for a spot and so could never be turned away.
A refusal can be caused by holds rather than by cars - a spot held by someone who has not arrived is still unavailable. Which of the two it was is not in this data.