The "perfect" suburban grid?

[tradephoric]

What suburban grid network would maximize vehicle capacity and reduce driver delay?  To answer this, more specific questions should be considered:

What speed limit should be chosen for a suburban grid? 
According to a report by the Texas Transportation Institute (Guidelines for Quantifying the Influence of Area Type and Other Factors on Saturation Flow Rate), the saturation flow rate at an intersection increases with speed.



However, it goes onto mention that lower speed limit streets are often associated with frequent driveways, high density roadside development, frequent pedestrian activity, and short segment lengths.  These factors may play more of a role on saturation flow rates than the speed limit itself.

What signal design should be chosen for a suburban grid?    
Two-phase signals that eliminate left-turn phases (often referred to as innovative intersections) can improve total capacity and maximize the "˜green band' of an intersection.  The big downside to most innovative intersection designs is that they inconvenience left-turning drivers.  However, a Town Center Intersection (TCI) enables direct left-turns while still operating as a simple two-phase signal.  When comparing the SYNCHRO models, it's apparent that the 2-phase TCI has a higher capacity than a conventional 4-phase signal (each model has the same traffic volumes & upstream geometry). 

There are other advantages to the TCI.  To cross a conventional intersection, pedestrians may have to cross 10 or more lanes of traffic (just look at examples of major intersections in Orlando).  With a TCI, pedestrians only cross 3 or 4 lanes of traffic at any given time.  Pedestrian crossings are much shorter which gives greater flexibility in regards to signal timings.  In addition, the wide flared out medians at the intersection provide development opportunities.

What cycle length should the signals along a suburban grid run?
Case for high cycle lengths:
Every time a signal changes phases, there is an associated "˜lost time'.  High cycle lengths reduce the number of times a signal will cycle per hour resulting in less lost time.  The useable green time, known as the effective green, will increase as the cycle length increases.  Below is a chart showing the effective green time for a variety of signal phases and cycle lengths (assuming 4 second of lost time per phase).



Case for short cycle lengths:
Research has suggested that the peak saturation flow rate of a signalized intersection occurs roughly 30 to 40 seconds into the phase green time.  After that, the saturation flow rate slowly tapers off.  Essentially, the saturation flow rate of an intersection isn't a constant value and changes based on the length of the phase green time (with longer phase times leading to lower saturation flow rates).  The chart below attempts to illustrate the concept.



A recent study, Revisiting the Cycle Length–Lost Time Question With Critical Lane Analysis — 2012, was performed testing the throughput of a saturated intersection running a variety of cycle lengths.  The intersection was a 3-phase intersection and had constant phase splits for each cycle length.  The following graph summarizes the results of the study:



The study found that a 100 second cycle length maximized the throughput of the intersection.  In addition, an 80 second cycle length had a higher throughput than a 135 second cycle length tested.  The study suggests that very high cycle lengths can lead to a decrease in throughput. 

How far should the traffic signals be spaced apart along a suburban grid?
The concept of the "˜resonant cycle' attempts to tie all these ideas together to start building the grid.  Put simply, a resonant cycle is a cycle length that achieves good two-way progression.  Below is a graph that shows the resonant cycle at varying signal spacing and speed limits: 



Looking at the graph, the ideal signal spacing for a 45 mile arterial running a 120 second cycle would be 3960 feet (0.75 miles).  This means signals that stop both directions of travel along the grid should be spaced 0.75 miles apart. 

As mentioned previously, the saturation flow rate increases as the speed increases.  Since suburban areas are less dense, higher speed limits are appropriate.  A 45 mph design speed is very common for suburban arterials.

Let's assume that a 2-phase signal has an optimum cycle length of 90 seconds under saturated conditions (this cycle would maximize throughput under saturated conditions).  Ideally, all traffic signals along the grid would be 2-phase signals to maximize capacity.  In reality, it's very likely that 3-phase and 4-phase signals would exist.  As the phases increase, the cycle length that maximizes throughput increases since the total "˜lost time' goes up.  Basically, the cycle length that maximizes the throughput for the "critical"  signal should be selected (which for a 4-phase signal would be around 120 seconds).  In addition, having slightly higher cycle lengths is preferred since this grid focuses on improving signal progression.  Higher cycle lengths lead to larger "˜green bands' which is a major goal of this grid network.

How should diagonal streets be set up to increase direct travel?
Diagonal streets allow for more direct travel for residents living in the NE, NW, SE, or SW suburbs of a city (who are commuting downtown).  In order to maintain progression, diagonal streets would cut through the middle of the major intersections.  The signal spacing along these diagonal streets would be 5600 feet.  In order to maintain dual progression along the diagonals, a 60 mph speed limit would be needed.  To achieve such a high speed limit, the diagonals would need to be limited access.

The suburban grid
So considering all the factors above, this is the suburban grid I came up with:



Highlights of the grid:

• Major intersections would be spaced 1.5 miles apart and have a speed limit of 45 mph.
• Town Center Intersections (TCIs) would be used exclusively at major intersections.
• Mid-block intersections spaced 0.75 miles apart.
• Michigan Lefts common at mid-block intersections.
• Major arterials would include wide medians, allowing for more flexible innovative intersection designs (such as the superstreet).
• Diagonal streets would be limited access and have a speed limit of 60 mph to maintain good dual progression.
• The grid could achieve dual progression at cycle lengths of 120 seconds (high cycle length) and 60 seconds (low cycle length).  A 60 second cycle length would help reduce driver delay in the middle of the night.  The rest of the time, the signals would run a 120 second cycle length.

[NE2]

Rearranging deck chairs on the Titanic...

[tradephoric]

Rearranging deck chairs on the Titanic...

Yeah, but Detroit has already sunk.  You pretty much have to ignore every rust belt city when making the claim that the suburbs are dying off.

Detroit Population:
1950 1,849,568
1960 1,670,144
1970 1,514,063
1980 1,203,368
1990 1,027,974
2000 951,270
2010 713,777
2012 701,475

Macomb County Population:
1950 184,961
1960 405,804
1970 625,309
1980 694,600
1990 717,400
2000 788,149
2010 840,978
2012 847,383

Oakland County Population:
1950 396,001
1960 690,259
1970 907,871
1980 1,011,793
1990 1,083,592
2000 1,194,156
2010 1,202,362
2012 1,220,657

[Brandon]

Rearranging deck chairs on the Titanic...

Yeah, but Detroit has already sunk.  You pretty much have to ignore every rust belt city when making the claim that the suburbs are dying off.

Ditto with Chicago.  Many "new urbanists" love to use Chicago as an example, but they're heading down Detroit's path pretty damn fast (massive debt - makes Detroit look good, flight, housing & business abandonment, etc).  The only part of Chicago doing well is the same part as in Detroit - downtown (here we call it "Daleyland").

[Zeffy]

Ditto with Chicago.  Many "new urbanists" love to use Chicago as an example, but they're heading down Detroit's path pretty damn fast (massive debt - makes Detroit look good, flight, housing & business abandonment, etc).  The only part of Chicago doing well is the same part as in Detroit - downtown (here we call it "Daleyland").

Newark is quickly heading in the same direction of Detroit. Trenton I think has a lot more time before it becomes that bad, and Camden almost looks like it's (very very very) slightly getting better since the new addition of the Camden County Police Department.

I haven't been to Chicago, but I definitely want to go there - but I can't say the same for Detroit.

[tradephoric]

Here's a few good examples of a Town Center Intersection:


https://maps.google.com/?ll=33.09823...04823&t=h&z=18


https://maps.google.com/?ll=39.760127,-104.890554&spn=0.001645,0.001725&t=h&z=19

Compare a gas station at the corner of a conventional intersection to a gas station built in the median.  It's apparent that a TCI can provide easy two-way access.  There are many situations at a conventional intersection where a driver turning left into a drive just before the intersection can block drivers wishing to make a left-turn at the intersection itself.  In fact, in the conventional intersection example, there is a raised median that prevents NB drivers from making a direct left into the gas station to avoid this very scenario.

https://maps.google.com/?ll=28.49395...00862&t=h&z=20