Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 43
43 CHAPTER SEVEN LESSONS LEARNED INTRODUCTION ­ It worked with NEMA controllers and cabinets ­ System is very flexible Each deployment can be different and challenging both to the ­ Side street venue dumps were dramatically improved vendor who installs the ATCS and the agency that operates ­ Centralized control, performance measures and maintains the system. Each deployment process is also a ­ Successful possible signal coordinating learning experience from which agencies can determine what ­ Spillback on Interstate reduced could have been done differently. This chapter identifies ­ Good early learning opportunities for ATCS major lessons learned through a series of questions answered ­ Appeared to handle delays about as well as promised. by interviewed ATCS users. First, ATCS users' positive and · Negative negative surprises from ATCS deployments were noted and ­ Black box--difficult to explain signal operation then the agencies provided their hindsight perspectives. Sub- ­ Video detection system a high maintenance effort sequent sections also present factors that caused several ATCS ­ Had hardware issue with communications, but it was shutdowns, problems that prevent further ATCS expansions, resolved and the potential for future ATCS deployments. ­ Lots of effort required to keep optimal performance ­ Steep learning curve USER PERSPECTIVES ­ System difficulty and lack of support ­ Hardware was discontinued by manufacturer Positive and Negative Surprises from ­ Much more sophisticated and complicated than told Adaptive Traffic Control Systems Deployments ­ Initial set-up costs and time taken ­ Amount of time needed for deployment The deployment of an ATCS is a challenging process for ­ Time and effort to debug central and field equipment both ATCS users and the vendors and consultants who per- ­ Importance of communications form system installation and integration. No two ATCS ­ Some phases not being served in early deployment deployments are the same. Idiosyncratic characteristics of ­ Design consultant made mistakes deployment site, traffic conditions, an agency's legacy hard- ­ There was a recurring issue with phase skipping. ware and software, and the institutional and cultural charac- teristics of the agency, make each ATCS deployment unique. For this reason each new ATCS deployment represents a People were mostly pleasantly surprised by ATCS's abil- new learning experience both for the agency and the vendor. ities to provide what was observed as "efficient operations" Although there is significant literature on ATCSs, people and to adjust to within-day and day-to-day traffic fluctua- who have not worked with an ATCS before tend to get a tions. The second important positive surprise related to a skewed picture about its abilities. This picture is sometimes systems ability to store traffic data that previously required intentionally skewed by ATCS vendors who are trying to sell much more difficult data collection efforts. Negative surprises intellectual properties and consulting services for an ATCS were mostly related to difficulties in learning how to operate to as many customers as possible. From that perspective, it is the system and hardware issues (mostly communications). interesting to discover what operational surprises are noticed There were also some complaints about operational features by agencies during their deployments. The following list of the systems (e.g., phases being skipped), which reflect a shows examples of positive and negative surprises. lack of hands-on expertise within the agency, availability of local technical support, or problems with hardware more · Positive than deficiencies in operational philosophies of their ATCSs. ­ How well it responded to changing traffic conditions ­ Short system response time to fluctuating demands There is a notion that some ATCS users jumped into ­ Effectiveness of the system installing an ATCS to replace an outdated fixed-time or iso- ­ Does adapt to daily commuter traffic lated traffic signal system. In such cases, the agencies won- ­ How well it worked; traffic data stored dered if the same operational benefits, which are achieved with ­ Moving special events traffic efficiently the ATCS, could be achieved with conventional actuated- ­ Better than expected results in before-and-after study coordinated traffic control.

OCR for page 43
44 Hindsight Perspectives to reduce installation costs--they turned out to be extremely unreliable To summarize what agencies have learned through their ­ Using 170 controllers, installing new communications ATCS deployment processes we asked them to answer what (fiber or twist pair) everywhere (tried to use existing they would, in hindsight, have done differently during the old twisted pair), starting smaller and then expand out. implementation of their ATCSs. Most of the responses indi- · Conduct a detailed pre-installation evaluation to esti- cated that greater emphasis might be given to local support, mate the operational benefits of an ATCS before decid- either through acquiring more in-house knowledge or by ing to implement an ATCS by ascertaining that technical support for their ATCS is avail- ­ Gathering more before-and-after data able locally. Better planning and preparation of the necessary ­ Not deploying an ATCS at areas with normal traffic infrastructure (communications and detection) are the next conditions important issues that most agencies would have done differ- ­ Running the operations with traffic signals under ently. Finally, some agencies recognized that they would need actuated coordination before deploying an ATCS. more information on the costs and benefits associated with installation of ATCS. Factors That Caused System Shutdowns The following is a list of the actions, summarized in four Lessons learned might also include instances where ATCSs major categories, which the interviewed ATCS users would were turned off. Although the few agencies that turned off have done, in hindsight, for their ATCS implementations. their ATCSs did not provide specific details to justify their decisions, the following is a list of major factors that influ- · Secure good local support from the vendor by enced those decisions: ­ Asking for dedicated support field staff; insisting on local vendor support · Improper detection layout and other operational prob- ­ Spending more time with the vendor's engineer(s) lems; ­ Keeping a more watchful eye on the contractor install- · Multiple simultaneous events: budget reductions, staff ing the system; design consultant had no expertise in reassignments, construction projects (resulting in sig- ATCS. Vendor, at first, did not supply the people with nificant removal of system detection); the most expertise--so the design consultant made · Operational problems; agency did not shut down the mistakes that had to be changed post-installation entire system, but it turned most of the ATCS signals to ­ Waiting for vendor-supported product rather then actuated-coordinated operations; deployed test application. · System incompatibility with ramp-metering where inte- · Improve the planning process to avoid in-house opera- gration of arterial and ramp operations was required; tional and institutional issues by and ­ Starting with a larger region · No operational benefits achieved; problems with hard- ­ Allocating more time for debugging ware and software. ­ Getting buy-in from operational staff and not just the steering committee and project management ­ Involving local staff at a much earlier stage and achiev- Problems Preventing Expansion ing full independence and expertise earlier as a result of Adaptive Traffic Control System ­ Doing it all in-house from the start ­ Appointing the necessary expertise to manage the The results from the survey indicated that the high cost of network. ATCS deployments is the most prohibitive factor to expanding · Prepare infrastructure (e.g., detection and communica- current ATCSs (50% of the cases). The second factor, by tions) for an ATCS deployment by importance, is the lack of traffic signal operations staff--a ­ Replacing all existing equipment--not retrofitting problem that can also be attributed to inadequate funding. ­ Using conventional loop detection or established non- Finally, some agencies reported that the operational ineffi- intrusive detection rather than the first release of a ciency of their ATCSs is the major reasons they have not new model of video detection expanded their systems. The following is a list of the major rea- ­ Testing (thoroughly) local control firmware and ease sons that interviewed ATCS agencies were prevented from of use for technicians expanding their systems spatially (to neighboring intersections) ­ Reviewing and planning for better communications; and temporally (to be used 24 hours a day/7 days a week): adding CCTVs for monitoring ­ Waiting for more current algorithms to be developed · Insufficient staff and funding to operate and maintain; before deployment; putting more thought into detec- · Poor communications between vendor and client; tor placement · Not cost-effective if volume fluctuations are insignificant, ­ Using inductive loop detection exclusively; micro- or where cycle lengths and splits are too constrained to wave detection was used for about half of the links meet operational objectives;