An important consideration in transportation planning is the efficiency of traffic flow. Traditionally, pedestrian traffic has not been given the same priority as vehicle traffic. Reducing motor vehicle delays at traffic signals may cause increased delay to pedestrians. In heavily traveled areas with high densities of pedestrian use, motor vehicle delay must be balanced with pedestrian delay. To encourage pedestrians to continue walking rather than returning to vehicles and increasing traffic congestion, travel-time costs to both groups must be incorporated in signalization system planning.

Purpose and Objectives of the Research

Robert B. Noland reported on the results of this study in "Pedestrian Travel Times and Motor Vehicle Traffic Signals." The paper appeared in the Transportation Research Board's Transportation Research Record No. 1553, Traffic Control Devices, Visibility, and Evaluations, published in 1996.

Urban planners are designing and planning for increased pedestrian traffic and reduced motor vehicle traffic. When walking, people generally do not travel more than one-half mile per trip. Considering this factor in planning suggests a focus on increased urban density and mixed land use. Locating employment and shopping areas together can increase pedestrian accessibility. Planners are being asked to allow for the development of small-scale communities that have employment, shopping, and residential areas all within walking distance. Providing traffic signals to improve vehicle traffic flow may increase pedestrian travel time and make walking less efficient.

Engineering requirements sometimes mandate certain street widths for increased high-speed vehicle traffic to maximize traffic flow. Crossing such streets causes major delays to pedestrians. This study considered the implications for new development design and might provide useful information for adapting existing developments to make walking a better option than in the past.

Quantifying Pedestrian Walk Times

The average walking speed is 3.1 miles per hour (mph) for men and 2.5 mph for women. For this study, the author assumed an average pedestrian speed in this range of 3 mph. A half-mile walk at this rate would take 10 minutes without stopping. A pedestrian walking for this distance across typical block lengths of one-eighth mile must cross four streets. If streets are narrow, traffic volume is light, and all intersections have four-way stops at which every vehicle yields the right-of-way to pedestrians, a person walking would experience a minimum of delay. There would still be some delay to look both ways before crossing to see if the street were clear.

This situation would require some degree of traffic calming. Vehicles might be restricted to 25 mph with stops at all intersections. If any restrictions were removed in the interest of improving traffic flow and speed, pedestrian travel times would be seriously increased.

An example assumed that traffic lights were installed at all intersections. Pedestrians would reach intersections at a green light only one-sixth of the time and would be delayed the rest of the time, provided they did not run to make the light and they did not walk against a red light (these behaviors would cost the pedestrian extra effort or jeopardize safety). Assuming pedestrians did not have to wait to cross when they reached the intersections, researchers calculated the average delay using both a simplified Poisson arrival distribution and a formula from the Highway Capacity Manual. The two formulas produced similar results.

Table 1 shows average delays for a variety of signal cycle lengths and pedestrian green phases.

If a pedestrian makes the half-mile trip with the delay times shown in Table 1, the total mean travel times increase, as shown in Table 2.

Motor Vehicle Delay

The Institute of Transportation Engineers (ITE) has calculated vehicle delays at signalized intersections and found that "when capacity is below about 75 percent of maximum flow, the green phases and cycle length have the most pronounced effect on average delay. . . . The calculations assume a saturation flow of 1.11 vehicles per second (veh/sec) (4,000 veh/hr) and an approach flow of 0.264 veh/sec (950 veh/hr)." Table 3 shows the average delay increase for vehicle traffic as the green phase is reduced.

Optimization of Signal Timing

If there is little pedestrian use in an area, short pedestrian cycles may be an option. In areas with a large amount of pedestrian traffic, however, signal timing should consider delays to people on foot. This may encourage systems that delay vehicle traffic longer than average or even ban motor vehicle traffic. The researcher conducted a literature review to determine the value of time to both motorists and pedestrians. He assumed that "waiting at a signal while walking is twice as onerous as waiting at a signal while driving."

The researcher also pointed out that "all travelers are at some point pedestrians," and he mentioned that transportation planners should consider that all people value their time as equally important.

The cost of travel delay to each group--motorists and pedestrians--was calculated by the average delay, the value of time, and the number of people using each type of travel. The research assumed vehicles to be single-occupant (SOVs). The presence of higher occupancy or transit vehicles may justify a longer delay for pedestrians. The researcher mentioned allowing transit vehicles to activate signals electronically. His analysis did not consider freight shipments and other commercial travel, or traffic and pedestrians traveling "at cross angles." These calculations produced the following results.

• When travel time costs of walking are twice those of driving, the cycle length is 60 seconds, and the ratio of pedestrians to vehicles is more than three, "aggregate travel time costs favor longer green phases for pedestrians."

• When travel time costs are equal, "signalization cycles favoring pedestrians give lower total costs when the pedestrian/automobile ratio is about 3 to 4."

• When travel time costs are equal and the cycle length is 120 seconds, "the most favorable pedestrian phase has the lowest total cost for all ratios greater than 1." This long cycle is economically sensible only if there is a large amount of vehicle traffic or very little pedestrian traffic.

• When pedestrian travel time costs are assumed to be half those of driving time costs, "the most favorable pedestrian phase still has the lowest costs when the ratio of pedestrians to automobiles is greater than 2."

Signal cycles that delay vehicles might increase traffic congestion and contribute to longer delays that may cause changes in trip timing, routes, and modes. One result of such changes could be an increase in pedestrian traffic.

Conclusions

1. In most cases, delay to pedestrians is not presently a consideration in the design of traffic signalization systems.
2. From an economic perspective, green phases for pedestrians should be shortened only when vehicle volumes are much greater than pedestrian volumes.
3. Where pedestrian volumes are high, the green phase should be lengthened for pedestrians--even if the value of vehicle and pedestrian travel times is equal.
4. Where pedestrian volumes are very high, closing some streets to vehicles during peak pedestrian travel times might be advisable.
5. A shorter signal cycle length (60 seconds) is more favorable to pedestrian traffic.
6. Reducing road widths would allow shorter signalization cycles.