CE 372 - Fundamentals of Transportation Engineering
(30 March 2007)

Questions and Answers
1. How are actuated signals used in combining consecutive signals?
Response:  The process of combining the operation of consecutive traffic signals is called coordination. While most downtown systems like Moscow operate in fixed time mode, many other systems operated in actuated-coordinated mode. There is usually one major street that is coordinated, and there is a fixed "green band" that provides this progress.  The actuated operation means that the side street phases may have variable durations.

2. What is the minimal acceptable LOS for an intersection.
Response: This is a good question and one in which there is wide spread disagreement in the profession.  Even though there is a common set of standards for level of service for intersection delay, what is acceptable varies from city to city. For example, we would find delays of 30 to 60 seconds on the upper end of acceptability in Moscow.  But if you were traveling in Seattle or Portland, delays of several minutes or more are not uncommon.  In fact, if you drive the I-5 corridor during rush hour, delays of 30 minutes or more are common.  Another view of this is contained in the AASHTO design guide.  Here, levels of service of C or D are commonly suggested as design standards.  But traffic engineers and travelers alike in larger cities would never expect this level of delay. There are just too many people traveling during rush hour to accommodate this the demand with this level of service.  Imagine intersections of ten lanes on each approach!  That is why most cities consider multi-modal solutions (including public transportation) as important parts of any solution that is developed.

3. When doing calculations, is there anyway to correct for the deviations in the initial assumptions we make?
Response: Yes.  Chapter 16, in the Highway Capacity Manual, is where the traffic analysis tools for analyzing signalized intersections are contained.  The chapter describes the basic model that we discussed on Friday, but then has a set of adjustment factors to account for relaxing the assumptions that we made in class.

4. In advance traffic courses, will we be using the simulation software you used in class today?
Response: Yes, in CE 474, we will be using VISSIM to evaluate signal timing designs that are developed during the class.  Also, in CE 372, you will use a version of VISSIM to learn more about signal timing principles.

5. Why did the state not want to put a stop light in by A&W?
Response: There are two issues: (1) the grade on the northbound approach that might be a problem for some trucks during the winter to stop for red and then accelerate for green, and (2) the spacing of the intersections (between the existing signals at Palouse River Drive and Sweet Avenue), making it difficult to design good coordination between the signals.  Intersection spacing is often the key problem in obtaining good coordination between adjacent signalized intersections.

6. Is the process of determining cycle length in Moscow the same that it would be in, say, L.A., California?
Response: The process is the same but the results are often different.  Why?  The demand in L.A., for example, is much higher than in many other cities, and the peak period can often last for several hours!  Often the problem is not progression or minimizing delay, but rather managing queues.  Interestingly, many traffic engineers believe that longer cycle lengths are the solution when volumes increase.  While longer cycle lengths are often needed to accommodate demand, the longer cycle lengths result in longer delays. Just for comparison, the cycle lengths in downtown Moscow range fro 70 to 80 seconds, while it is not uncommon to find cycle lengths of 200 to 300 seconds.

7. Can the simulators sense small movements such as pedestrians and scooters?
Response:  First a distinction: simulators don't sense movements, rather they simulate the response of users (pedestrians, car drivers, etc) to various conditions such as traffic flow levels, geometric conditions, and control conditions.  Simulators model what we build into the model.  With regard to simulators, models like VISSIM to consider pedestrian movements (and we can observe pedestrians in the animation). 

8. I would like a better understand of the two perspectives. Are you talking as both perspectives being from the same person, or is one from the engineer and one from a motorist point of view?
Response: Here are some additional thoughts on the perspectives that I was talking about during class.  Too often, when we are evaluating the operation of any system (not just a transportation system) we take the view of an engineer, that is our own perspective.  For a transportation system, we will use performance measures such as volume/capacity ratio or queue length.  But what does the volume/capacity ratio mean to a traveler?  In 1985, the committee that oversees the Highway Capacity Manual decided that all measures of effectiveness (or MOEs) for each kind of traffic facility (freeways, two lane rural highways, signalized intersections, etc) would be from the user perspective.  So, while a traffic engineer might find value in evaluating the volume/capacity ratios for a signalized intersection, all measures of level of service would be using MOEs based on the user perspective.  That is why delay is used for intersections, since we as travelers can relate directly to how long we've been stopped for the red light.  Try to assess the volume/capacity ratio when you are stuck at red light!

9. On an intersection like the one we are looking at, how do you accommodate for pedestrians with out stop lights. At this intersection I find it hard to get across. Could the intersection be marked better to warn people that a pedestrian crossing is ahead?
Response: This is an on-going challenge for both traffic engineers and citizens, and here is an example from Moscow.  About a year ago, the city of Moscow and the Idaho Transportation Department responded to concerns about crossing State Highway 8 between downtown and Palouse Mall. The problem has been exacerbated by the construction of new apartments on A Street, greatly increasing the number of people crossing the highway.  The solution developed by the city and ITD was to add two specific cross walks on State Highway 8, and to design marking and feeders from campus to these crosswalks.  There are also new warning signs and eventually there will be more warning lights.  This has helped pedestrians, but it has resulted in an increase in the number of rear end vehicle crashes on the highway. There is never a solution without some consequences!

10. How do you account for the change in level of service if the queue doesn't clear within the green time?
Response: We would use the same approach that we used for the model that we developed in class.  The difference is that we would show the queue extending into the next red interval and then assessing the following green interval.  The delay increases significantly if the queue doesn't clear during green, as we can see in the area below.

Queue clears during green:

Queue doesn't clear during green and delay (as shown by the area in the polygon) is larger.

11. How are actuated signals and different traffic conditions accounted for?
Response: Chapter 16 of the Highway Capacity Manual has procedures to account for the variability in the green times and cycle lengths that we observe in actuated controllers.  If you mean by traffic conditions the volume or flow rates, this is explicitly included as one of the inputs to the models.

12. For me the intersection at Styner, Lauder, and HWY 95 has a problem with visibility. I can't see well when trying to make a left from Lauder. How does visibility play a factor in LOS or other traffic operation analysis?
Response: Visibility is often not accounted for by various factors in models like the Highway Capacity Manual, even though it should be.  A surrogate factor for visibility is often reduced speed, which is included in some capacity models.  There is a new project funded in part by the Federal Highway Administration and the states to develop a new Highway Safety Manual, as a complement to the Highway Capacity Manual. 

13. You mentioned that there are many assumptions made when deriving the equations we went into today, how do we counteract or take into account those assumptions?
Response: These assumptions are relaxed with the addition of adjustment factors in the models such as those included in the Highway Capacity Manual.

14. How can we factor in more variability when analyzing an intersection?
Response: If you mean consider more factors than we discussed in class, these additional factors are included in most models.

15. How does all of this tie into the overall design of an unsignalized intersection?
Response: We will discuss the model for analyzing capacity and delay at an unsignalized intersection on Monday.  The benefit of having models for both types of intersections is that we can determine which will perform better.

16. Is a simulation normal practice for most large scale transportation problems? How do you decide to use it?
Response: As I mentioned in class, the type of model used depends on the kind of question that you are trying to answer.  If you want to know that number of lanes required for a given set of volume conditions, the Highway Capacity Manual is a good tool.  If you want to test details on signal timing parameters or if congestion exists (volume exceeds capacity), then simulation models like VISSIM are often used. The more complex problem (which usually means that more money is being spent on the solution) the more extensive the analysis (and modeling).

17. How do you account for incorrect assumptions in the equations derived in class?
Response: Change "incorrect" to "limiting" or "initial".  For example, if volume is not uniform but is more random, there are adjustment factors to use.  If signal timing parameters vary (as in actuated signals), there are other adjustment factors.

18. What level of service are intersections typically designed for?
Response: See question 2 above.

19. When does the level of service factor for the few out weigh the level of service considerations for the many?
Response: For a signalized intersection, for example, level of service (based on delay) is computed for individual movements, approaches, and for the intersection as a whole.  And if we are looking at an arterial, we can compute the level of service for the arterial.  There is no easy answer to your question, but in all design decisions these trade-offs are considered.

20. How often are intersection performances usually analyzed?
Response: Not often enough. Land use patterns are constantly changing and thus travel volumes change as well.  In areas in which growth is high, these analyses should be performed regularly.  Unfortunately, they are not. Limited funding is usually the reason.

21. If level of service is based on average delay, are there extreme cases when one road has very little traffic crossing a heavily trafficked road, where the smaller road should be given the right of way even if it means delaying many travelers on the large road?
Response: Intersection design and signal timing is a truly democratic process. The higher the volume on one approach, the more green time is allocated to that phase.  In the extreme that you describe, it is important to give most of the green time to the main street, but some green time must be allocated to the side street.

22. What happens if all of our assumptions don't reflect reality (which of course they don't)? How do such alterations affect the results of these HCM models? Relating to this, how closely can we rely on these models for basing decisions, such as whether the intersection of Styner and US 95 should be signalized or a two-way stop?
Response: In addition to the responses to the above questions, the models must be calibrated to reflect local conditions.  The calibration process is important to make sure that assumptions (such as saturation headway, for example) reflect driver behavior and travel patterns in a particular area.

23. Why are fixed signals used if actuated ones respond better to changes in traffic?
Response: Fixed time signals are used most often in areas in which travel volumes are relatively stable and when the objective is to have progression on one or both streets.

24. How do the models we developed compare to actual traffic statistics?
Response: If the models are calibrated to local conditions, the comparisons are good.  But when we are dealing with models of the real world, they are never exact.  One of the big challenges for students first learning to use models is to gauge their expectations.  We can learn a lot about how a traffic system is likely to perform by using models, but it won't tell us everything.  In the end, we must learn to use the results from the models with judgment.  That is always the biggest challenge!

25. Could you maybe go over finding the average delay using the diagrams, I'm just a bit confused?
Response: I'll provide a brief review in class on Monday.

26. Are there other ways to compute average delay if all of the parameters we used aren't available?
Response: We can measure it in the field directly. But if we are using a model (like we did in class on Friday) we need at least the variables that we discussed.

27. I am possibly interested in taking your traffic systems design class and was wondering how the class was structured.
Response: The class will focus on learning more about signal timing for actuated traffic control systems.  Students work in teams and have two design projects to complete during the semester.

28. What programs do you utilize in your CE 474 class?
Response: VISSIM and the HCS.

29. Statistically, are there more ped/bike accidents at light or sign controlled intersections? And is it safer, again statistically, to bike in a crosswalk or a car lane?
Response: I don't know the differences off hand between crash rates at signalized or stop controlled intersections. The FHWA safety web site (http://safety.fhwa.dot.gov) might have more information.

30. What is the deal with Eagle Rd. in Boise?
Response: I'm assuming that you mean the high volumes and delays on Eagle?  If so, here is a quick response.  First, the responsibility for traffic engineering in Boise rests with both Ada County Highway District (ACHD) and the Idaho Transportation Department (ITD).  Both agencies have participated in a series of design reviews for new projects along Eagle Road during the past several years.  While I haven't followed the projects in the past two years, I do know that there were a series of agency and public workshops held in 2004 in which all of the stakeholders worked together to find both a geometric design and signal timing plan that could handle the growing traffic demand along Eagle Road. There is also consideration of a new north-south arterial that could provide some relief.  But the biggest problem is the need to recognize the land use decisions (i.e., more development) must be made jointly with transportation investment decisions.  If not, then the result is the kind of delay that you experience today along Eagle Road.

31. How is the level of service affected when there are many signs near the intersection?
Response: If you mean do more signs affect visibility and driver confusion?  If so, this is not directly taken into account in computing level of service.

32. Who calculated signaling for Palouse Mall before its construction?
Response: This was done by the Idaho Transportation Department, since they are responsible for State Highway 8.

33. Who is the city of Moscow's transportation engineer?
Response: Moscow has a public works director (Les MacDonald) who is responsible for all public works including transportation.  There is also a Transportation Commission (chaired by Walter Steed) that oversees transportation policy for the city.  Here is the web site for the Moscow Transportation Commission: http://www.moscow.id.us/commissions/transportation.asp.

34. How do you calculate delay when signals are in series?
Response: When traffic signals are interconnected, there is the potential to lower delay through signal coordination.  The same basic method that we talked about during class is used, except for an adjustment factor that accounts for when platoons from the upstream signal arrive.  If they are more likely to arrive during green, the delay is less than our delay model would predict.

35. Does the HCM specify delay length?
Response: If you mean does the HCM specify a range of delay ranges for each level of service, then the answer is yes.

36. 2 things I still have questions about: How do you determine which analysis program is the best for the intersection you are analyzing?
Response: The analysis program to use depends on the nature of the problem that you are analyzing.  See the response to question 16 above.

37. How do you determine the saturation flow rate for the intersection you are analyzing?
Response: Basic parameters such as saturation flow rate must be measured in the field.  The saturation headway is measured by determining the time interval between the departure of two consecutive vehicles departing from the stop line at the beginning of the green interval.  Because of the "lost time" (the time that it takes the first few drivers to respond to the change from red to green), the headways are measured beginning with the fourth vehicle in the queue.

38. How do intersections accommodate bicycles? Bicycles do not trigger the actuators.
Response: You are correct that many detectors don't seem to trigger bikes.  Most traffic engineers are aware of the issue and in situations where there are a number of bikes, the sensitivity of the loop detectors are tuned to detect bikes.

39. I was wondering how much of an impact on signal efficiency our assumptions about traffic make.
Response: Signal efficiency is important.  It is sometimes defined in terms of the amount of wasted or underutilized time at the end of green.  While inefficient operations are not desirable, knowing what the green times are are key in the accuracy of the models

40. I was also wondering when transportation engineering, specifically traffic engineering, became crucial for America.
Response.  Traffic engineering became recognized as a profession in the 1940's and 1950's.  It has evolved greatly during the past fifty years, going from a civil engineering discipline (where it is still housed in most academic departments at universities) to a broader multidisciplinary effort today.  In fact, in our transportation research at the University of Idaho, we commonly have teams of civil engineers working with electrical, computer, and mechanical engineers on projects.  One discipline, known as Intelligent Transportation Systems, requires computing, communications, networking, and other traditional electrical engineering expertise.  See this link to ITS on the U.S. Department of Transportation web site: http://www.its.dot.gov/index.htm

41. How do our assumptions effect the model?
Response: The assumptions that we made in class on Friday are reasonable and give reasonable results.  But calibrating our model to field conditions that are present in our study area further improve the accuracy of our estimates.

42. How long have you been working in this area of Civil Engineering?
Response: First answer: too long!  Second answer: since 1972 in transportation engineering, and since 1986 at the UI.

43. In the equation, c=s(g/C), I know the small c is capacity, but I noticed that the other was a capital C. What is that variable?
Response: The capital C is cycle length.

44. Where exactly was the Hawkins land where they want to put a Home Depot.
Response: There is both a Home Depot and Lowes planned for the area, one is a part of the Hawkins development and the other is on the Idaho side, just west of the Wal-Mart development.

45. How are intersections accommodated for things that don't trigger the actuators, like bicycles?
Response: See #38 above.

46. Can the simulator simulate pedestrians?
Response: Yes.

47. How is the queue that occurs during the green time get accounted for?
Response:

48. What are some of the proposed solutions to the intersection of Lauder and 95?

49. How do you model actual arrival?

50. Are the models fairly accurate?

51. Can the pedestrians be simulated in the simulation?

Things of Interest
1. The intricacy of the simulation model you showed us, and how good the simulation appeared, visually, on the screen.
2. Using the graphical solutions to solve for delay.
3. Simulation software.
4. Using graphics to determine delays and headway using equations. More emphasis should be placed on this.
5. I liked the talk on the traffic program that incorporated the whole city of Moscow.
6. Real world topic such as the Wal-Mart topic is interesting to talk about.
7. It's interesting how data can be selected on a certain day (sometimes over several different periods) and effectively determine an adequate cycle time.
8. It's interesting how we are allowed to make certain assumptions, such as uniform flow and driver attentiveness (pertaining to headway), to help us determine cycle length.
9. The analysis program.
10. Learning about the impact of Wal-Mart and the impact on the community.
11. I learned how to calculate total and average delay.
12. I learned how to visualize or graphically present relationships between the different characteristics of a signalized intersection, such as time and arrival rate, or queue length and time.
13. I really enjoyed the 3-D simulation.
14. I found interesting (although still not completely clear) the interrelationship of the graphs we looked at.
15. Microsimulation is a good way to visually represent how the overall design of signalized system.
16. Seeing the graphs for the concepts we were discussing was helpful in understanding everything better.
17. Class seemed to go a little fast today, it was hard to write down any notes.
18. I really liked the focus on some of the recent events around Moscow.
20. VISSIM has the capability to create impressive traffic simulations.
21. The Palouse Mall is 300,000 sq ft.
22. The VISSIM model seemed very interesting.
23. The evaluation topics, when considering to put in a new business like Wal-Mart such as increased traffic flows, new pedestrian paths and bike lane, etc.
24. I found the discussion about Wal-Mart very interesting. I'd like to know more in the future about the exact effects that putting in such a large facility would have. I think that I would enjoy performing and learning about the analysis that engineers would have and likely did perform when considering such a proposal.
25. I enjoyed the micro-simulation. Learning to develop such simulations would be of interest to me (which it sounds like we are going to do in lab next week). The calculations and HCM models found in class seem reasonably straight forward, but I'd like to learn more about the complicated situations that deal with much more in depth measures.
26. Simulation of downtown Moscow traffic.
27. Relevance of issue being discussed (Lauder/Styner and US95 intersection)
28. The first program shown was very interesting and appeared to be a very useful tool, if you know how to operate it.
29. Using the diagrams to compute the average delay was a new method for me, although I didn't totally grasp the process.
30. Real world applications always seem to put classes in perspective. When talking about the Wal-Mart issue I realized how interconnected all the CE disciplines are, and how it affects the public’s well being.
31. The simulation software I thought was very festinating. I wouldn’t mind discussing and seeing other applications of this kinds of technology.
32. Although not directly related to transportation, I was very interested in the Hawkins development and it's impact on the community.
33. Additionally, the 3d simulation of Moscow traffic was interesting in it's potential to demonstrate proposed upgrades to the public.
34. More about LOS.
35. The new Wal-Mart is going to be greatly oversized like everything else in America.
36. VISSIM has large capabilities.
37. Palouse Mall is 300,000 ft^2.
38. How to calculate delay for an intersection.
39. That there are 3D traffic modeling computer programs.
40. I found the concept of saturated flow rather than just flow to be interesting and more helpful. My internship at Ada County Highway District this past summer and this class have showed me that a great deal of what we learn in the classroom show up in the workplace.
41. The time/space relationship.
42. The size of the proposed Wal-Mart and the implications.
43. The simulation was awesome. I really liked how you put it on a Google earth background.
44. The whole development of a Wal-Mart or Home Depot, I never realized how much that would affect the traffic and what the city would have to do to accommodate them. Very interesting.
45. The Wal-Mart Dilemma
46. The simulation of the downtown corridor.
47. The simulator was great! It was interesting to study intersections near mine!

The problems and concerns that are associated with a new development. Computer simulation models.