Improving Intelligibility of Airport Terminal Public Address Systems (2017)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

7 2.1 Introduction To assess how various stakeholders perceive intelligibility of PA announcements in airport terminals, the research team developed a questionnaire about respondent understanding of the factors involved in speech intelligibility and whether respondents believe speech intelligibility is a widespread problem. Given that respondents may not be familiar with specific terminology, many questions included multiple-choice answers to guide the responses and lend consistency. In many cases, respondents also could choose to provide additional comments or select “other” as a response. The questionnaire targeted airline industry people who have a role in the operation of the airports about communications transmitted over the PA system (i.e., managers, facilities per- sonnel, passenger services, IT and security, and related consultants and vendors). These results are only a sample of the industry—with only 66 responses, these results have a 10% margin of error to achieve a 90% confidence level for a population of about 400 airports and 100 airlines. All survey respondents were asked some of the same questions, and those common questions and answers are summarized in the next section. To understand how data might be collected from passengers, a pilot passenger survey was conducted at one airport. This effort sought to determine the feasibility of such a survey, as well as whether this type of survey could be conducted on a larger scale and how reliable and mean- ingful the data would be. 2.2 Overall Industry Perspectives A total of 66 individuals responded to the online survey. These individuals represented 38 airports, 5 airlines, and 21 consultants, vendors, and trade representatives. The median years of experience in the industry across all respondents was 25 years. There were five questions about PA systems in common for all respondents, and the results are summarized in Table 2-1. The following figures illustrate the questions with multiple options, and, for some questions, the other minor responses and gradations in response. For the questions in Figures 2-1 and 2-2, many respondents gave no response or were unable to respond to the questions. The question in Figure 2-3 allowed multiple selections, and the Other category was selected by 11% of the respondents. The additional factors offered under Other included microphone selection, user training, poor acoustical design, poor commissioning, gate areas in close proximity, number of announcements. C h a p t e r 2 Industry and Passenger Perspectives

8 Improving Intelligibility of airport terminal public address Systems Question Primary Response(s) Other Substantial Responses Are poor PA systems a widespread problem? 55% agree 35% no response* 10% disagree Should PA systems be improved? 46% agree 27% no response* 24% disagree What negatively affects the speech intelligibility of PA systems (multi-select)? Room acoustics (68%) Background noise (67%) PA electronics/design (45%) PA layout (48%) Announcement quality (50%) Areas where the PA system could be improved (multi-select)? Gate areas (64%) Concourse (68%) Departures (41%) Ticketing (41%) Baggage claim (45%) Curbside areas (47%) Source of information (multi-select)? Personal observation and experience (95%) Discussions with other facilities (55%) Discussions with airlines staff (44%) Customer feedback (24%) * These respondents did not have the opportunity to answer these questions. Table 2-1. Industry survey, primary findings. No response I strongly disagree I disagree Other (neutral) I agree I strongly agree Figure 2-1. Question: Are poor PA systems a widespread problem? (66 respondents)

No response I strongly disagree I disagree Other (neutral) I agree I strongly agree Figure 2-2. Question: Should PA systems be improved? (66 respondents) I don’t know/No response Other Poor quality PA system design or electronics (e.g., cheap speakers with limited or no adjustment opons) Poor quality announcement recording/ Poor user training (e.g., bad enunciaon or poor microphone technique) Poor speaker layout/design (e.g., too few speakers, too far from passengers) High background noise (e.g., close to curbside or near retail and mechanical HVAC equipment) High room reverberaon/poor acouscs (e.g., linoleum/le floor areas with glass and liƒle or no acouscal absorpon) Figure 2-3. Question: What negatively affects the speech intelligibility of PA systems? (66 respondents)

10 Improving Intelligibility of airport terminal public address Systems The question in Figure 2-4 allowed multiple selections, and the Other category was selected by 9% of the respondents. The additional factors offered under Other included restrooms and con- cession areas, back of house, highly reverberant areas with hard surfaces, and separate/remote rental car customer service areas. The question in Figure 2-5 allowed multiple selections, and the Other category was selected by 6% of the respondents. The additional sources of information offered under Other included passenger feedback, building tenants, and previous experience in IT management. 2.3 Airports and Airport-Based Staff 2.3.1 Summary Airports from across the United States and Canada provided input to this research (see Fig- ure 2-6). The responses received from two Canadian airports were consistent with the results from the U.S. airports. Of the 66 total responses, 38 airports are represented by 46 respondents. Table 2-2 and Figures 2-7 through Figure 2-12 summarize the results. 2.3.2 Detailed Results For the information in Figure 2-7, in addition to Spanish, French and Chinese, the Other lan- guages were German and Japanese. Some respondents used this field to indicate that they only use non-English languages as needed (e.g., for specific flights). I don’t know/ No response Immigraon and customs Other Gate areas Concourse Departures hall Curbside areas Arrivals hall Tickeng Baggage claim Figure 2-4. Question: In what areas could the PA system benefit from improvements? (66 respondents)

Industry and passenger perspectives 11 I don’t know/ No response Other Discussions with facilies and operaons staff (airport facilies) Discussions with operaons staff (airlines) Personal observaon and experience Customer surveys/ Feedback Figure 2-5. Question: What informs your understanding of this issue? (66 respondents) Figure 2-6. Questionnaire responses, geographic distribution— United States.

Question Popular Response Other Substantial Responses What type of service? 71% domestic and international 29% domestic only Non-English announcements 66% no or none 34% yes Non-English languages* (multi-select) Spanish (34%) French (11%) Chinese (5%) Non-English options* (multi-select) Prerecorded or automated (39%) Language specialists (29%) Announcement quality controls PA system design criteria** 75% no/no response 25% yes Acoustical design criteria** 84% no/no response 16% yes Speech intelligibility design criteria** 93% no/no response 7% yes Positive factors Good speaker layout design (72%) Good system design or electronics (61%) Low background noise (43%) Good room acoustics (43%) Good announcements/ recording quality (39%) Negative factors High background noise (58%) Poor PA design (45%) Poor room acoustics (50%) Poor PA equipment (50%) Poor announcements/ recording quality (42%) Areas at your airport that are generally good Gate areas (71%) Baggage claim (53%) Ticketing (50%) Concourse (50%) * These responses are only for those airports providing non-English announcements. ** The remaining respondents either responded “Unknown” or gave no response. Areas at your airport that are poor Curbside (39%) Ticketing (34%) Departures halls (29%) Table 2-2. Industry survey, responses from airport-based staff. No response Other None Chinese Spanish French Figure 2-7. Question: Which languages other than English are used for standard message announcements? (38 airports represented)

Use language specialists to provide prerecorded or live messages Use prerecorded or automated messages Other Figure 2-8. Options for non-English announcements. (23 responses) Good quality PA system design or electronics (e.g., ambient sensing, local adjustment controls, etc.) Good speaker layout/ design (e.g., enough loudspeakers to reach passengers) Good quality announcement recording/user training (e.g., good enuncia
on and microphone technique) Low background noise (e.g., away from mechanical HVAC equipment) Low room reverbera
on/ good acous
cs (e.g., carpeted floor, acous
cal ceiling
les) Figure 2-9. Question for airport-based respondents only: What are positive factors for PA system speech intelligibility? (38 airports represented)

Poor quality PA system design or electronics (e.g., cheap speakers with limited or no adjustment opons) Poor quality announcement recording/ Poor user training (e.g., bad enunciaon or poor microphone technique) Poor speaker layout/design (e.g., too few speakers, too far from passengers) High background noise (e.g., close to curbside or near retail and mechanical HVAC equipment) High room reverberaon/poor acouscs (e.g., linoleum/le floor areas with glass and li le or no acouscal absorpon) Figure 2-10. Question: At your airport, what are negative factors for PA system speech intelligibility? (38 airports represented) No response Other Gate areas Concourse Departures halls Curbside Areas Arrivals halls Tickeng Baggage claim Figure 2-11. Question: At your airport, what areas are good? (38 airports represented)

Industry and passenger perspectives 15 For the information in Figure 2-8, of those 14 airports providing non-English language announcements, the Other techniques to control the quality of these announcements included handled by Public Affairs, professional recordings of customer-service–approved text, reviewed by IT department, professionally produced and reviewed by system engineer, and reviewed and approved by airport director and marketing manager. Some of the questions were refocused to the specific environment at each terminal where airport-based–staff work (Figures 2-9 through 2-12). 2.4 Airlines Of the 66 total respondents, 13 were affiliated with airlines or the airline industry, including 6 from 2 major U.S. passenger carriers, and 3 from other, smaller air carriers. 2.5 Consultants Of the 66 total respondents, 12 were consultants for design and planning firms, and 6 were vendors/installers representing 3 firms, including 1 major vendor of PA systems and 2 firms that provide installation or integration of PA systems. 2.6 Passengers A passenger survey was tested at one of the airports to determine the feasibility of such a survey, whether this type of survey could be conducted on a larger scale, and how reliable and meaning- ful the data would be. The data obtained from the test survey was useful in obtaining feedback No response Other Gate areas Concourse Departures halls Curbside areas Arrivals halls Tickeng Baggage claim Figure 2-12. Question: At your airport, what areas are poor? (38 airports represented)

16 Improving Intelligibility of airport terminal public address Systems from passengers. Speech intelligibility measurements were also conducted in the areas where the passenger surveys were conducted. The aim of the passenger survey was to gain understanding and insight into how passengers hear, understand, and perceive different kinds of PA announcements when traveling through different airport touchpoints. The passenger survey was developed to focus on human factors, and, in asking passengers what they have heard and what issues they have encountered with PA announcements, endeavored to gain a passenger’s perspective of PA intelligibility that could be used to inform the acoustical study. (See Chapter 5 for further discussion on Human Factors and their role in this topic.) The passenger survey involved engaging directly with airport passengers to obtain their views on PA announcements in general, how they hear those PA announcements, what content they listen to/perceive, what issues they have encountered, and what factors may affect their abil- ity to perceive announcements correctly. A total of 76 passengers were surveyed using trained ergonomists with experience in survey design and collection. A copy of the survey questions and result summary is included in Appendix B. A total of 43 passengers were aware of an announcement just prior to the survey. Of these, four indicated that the announcement was muffled or unintelligible. Of the remaining respon- dents who heard enough to understand the announcement: • Fewer than 10% felt that the announcement was relevant to their journey. • About 18% did not understand the content or meaning of the message. • Reasons cited for not hearing the message well included high background noise, poor sound quality (echoes/distortion), message spoken too quickly, announcement volume too low, out- side noise, message not spoken clearly enough, and audio clutter/multiple messages. 2.7 Conclusions The major conclusions from the industry survey are as follows: • Although 55% agreed that poor speech intelligibility from airport PA systems is a widespread problem, only 46% agreed that existing PA systems require improvement. • Across all industry responses, most respondents indicated that high background noise and room acoustics were the most important negative factors on speech intelligibility of PA system announcements, with other factors relating to the design of the PA, electronics, and layout receiving responses from 45% to 48% of the respondents, and poor-quality announcements receiving 50%. • Airport-based staff were relatively even in their expectations of what hinders speech intel- ligibility of PA system announcements, with high background noise receiving 50% of the responses. • Gate areas and concourses received the most votes for areas of the airport that could be improved, although airport and airport-based staff generally thought that the gate areas in their own air- ports are satisfactory. This dichotomy is likely because gate areas are the places where passengers are most anxious about announcements related to boarding, delays, upgrades, and so forth and these are considerations that airport staff do not have. • Most opinions reflected personal experience or interaction with passengers or airport staff. Nearly one-quarter of the respondents had information from passengers. • Of the airports surveyed, almost three-quarters provide international service. • More than one-third of the airports broadcast standard announcements in a non-English language, and some technique or process is used to control the content or quality of the announcements.

Industry and passenger perspectives 17 • Of the airports surveyed, one-quarter of the airports were aware of specific PA system design criteria. • Most of the airport-based staff believe that PA system design or installation is one of the most important factors for speech intelligibility, with room acoustics and background noise following announcement quality. Major conclusions from the pilot passenger survey are as follows: • The questionnaire results can vary widely across airports and different areas of the airport; for example, passengers in the gate area tend to be more aware of announcements than those in other areas of the airport. • Passengers have different needs based on the phase of their journey (see Chapter 5 re human factors). • Many passengers rely on sources of information other than PA system announcements. This may be true for various reasons, one of which could be that they are most likely not to under- stand the PA announcements. • Based on the results obtained, the questionnaire used for the passenger survey is a useful starting point for airports to develop and implement their own questionnaires.