Cross-town Improvement Project Evaluation
4.0 Delphi Study Results
In order to further gauge C-TIP benefits, including more subjective criteria such as the importance of C-TIP software features to users, a targeted Delphi assessment was conducted. Delphi is an analytical method that attempts to quantify expert opinion on a given topic through an iterative polling process. The Delphi approach relies on a survey panel of subject matter experts, who are polled to obtain their opinions about a given topic. Through iterative survey rounds, the panelists’ responses are collected and shared anonymously until no further movement towards consensus is evident. The result is the Delphi panel’s collective opinion on a value or range of values. Although it does not represent a statistically valid sample, the Delphi process does provide a snapshot summary of expert opinion on the topic, with reasoning to support differing and similar viewpoints. This is particularly useful for capturing information which can’t be easily quantified.
The Delphi assessment focused on evaluating the importance of key C-TIP features, and on estimating the potential travel time, bobtail reduction, and other benefits that might be associated with a theoretical Chicago deployment of C-TIP technologies. This was done due to the low participation rate found in Kansas City; since Chicago is the largest intermodal hub in North America, it stands to reason that C-TIP would have greater benefits there and may attract more interest.
The Delphi process was conducted in two rounds via telephone interviews, e-mail, and fax. Questions revolved around the importance of C-TIP features/functions to users; estimated time savings from RTTM and DRG in a theoretical Chicago deployment; estimated bobtail and idle reduction from a Chicago deployment; potential railroad benefits from C-TIP; and the ability of C-TIP to reduce manual processes such as track and trace and dispatching. The first and second round survey instruments are provided in Appendix C.
Fifteen interviews were conducted in the first round, representing the viewpoints of 17 individuals who are experts in the field of intermodal transportation. A list of Delphi panelists is provided in Table 4.1 . Respondents included representatives from four Class I railroads, three drayage companies, one terminal operator, one steamship line, and one independent consultant. Some respondents had been involved previously with C-TIP in a conceptual or operational capacity and were, therefore, already very familiar with it; others were new to the project. CS developed a PowerPoint presentation to familiarize participants with C-TIP and the Delphi process; this presentation was sent out to each respondent prior to first round interviews.
Note: Delphi panelists were selected based on recommendations from SAIC and RMI.
Upon completion of the first round of interviews, all responses were entered into a spreadsheet for analysis. Questions involving Likert scale responses or ranges of values (e.g., percentage of travel time saved) were coded according to the values defined in the Delphi instrument. Qualitative responses were typed manually into the spreadsheet for comparison across panelists.
A second round questionnaire was then developed for those questions where there was considerable variation of opinion. The second round questions were targeted at those respondents whose answers were in the minority for that question. The new questions presented the majority viewpoint, with reasoning, and respondents were offered the chance to amend their answers or provide further defense for their original opinion. Overall, the second Delphi round did show considerable movement towards consensus on questions where there had been a wide range of responses.
Importance of Features
The first group of Delphi questions attempted to gauge the relative importance of different C-TIP functionalities to potential users. Respondents were asked to rate various features of C-TIP on a scale of 1 to 5, with 1 being not at all important and 5 being extremely important. The results are presented in Table 4.2. The shaded cells represent the group consensus values. Most of the panelists felt that C-TIP’s congestion/traffic avoidance, minimizing terminal delay, and load notification features were either ‘very important’ or ‘extremely important.’ Overall, this result lends credence to the idea that there is an opportunity to improve the flow of intermodal freight with a technology like C-TIP that seeks to address these issues.
There appears to be agreement that C-TIP routing features (RTTM/DRG) are slightly more important to users than minimizing terminal delay or automatic notification of load availability. Eighty percent of the panelists rated ‘avoiding recurring congestion’ and ‘avoiding traffic incidents’ as either ‘very important’ or ‘extremely important.’ Many respondents felt that delays in intermodal terminals already were minimal (the railroads in particular stated that they monitor this metric very closely and have worked hard to reduce turn times over the last few years). Meanwhile some panelists expressed skepticism about the feasibility of a common dispatch platform that would enable an automatic load notification function; although many acknowledged the benefit this could bring, they also noted practical obstacles to making it work, including reluctance to turn over the dispatch function to an outside entity.
Source: CS analysis of C-TIP Delphi responses.
Travel Time Savings
The next set of Delphi questions sought to estimate the per-trip travel time savings that might be achieved in a theoretical Chicago deployment of C-TIP. This question was asked separately for both RTTM (initial route guidance) and DRG (en-route redirection). As shown in Table 4.3, the group consensus fell into the five percent to 10 percent range for both RTTM and DRG, though there also were several respondents who felt RTTM could achieve time savings of 10-15 percent per trip. There was broad agreement that the potential is greater for initial route guidance (assuming alternate routes are viable for trucks), because accidents or other events that cause non-recurring congestion are random by nature and therefore aren’t likely to affect as many trips. Overall, respondents felt that there is likely to be some sort of delay (accident, traffic, or other) on a proportion of trips regardless, and a route recommendation (either before starting the trip or en-route) would help mitigate this. The savings would therefore be a function of the level of the opportunity, which in a city like Chicago could be substantial.
Source: CS analysis of C-TIP Delphi responses.
Bobtails
Another set of questions asked the panelists to provide order-of-magnitude estimates of daily bobtail volumes in Chicago, and the potential for C-TIP to reduce the number of bobtail trips. As shown in Table 4.4, six panelists felt there are between 301 and 500 daily bobtail trips in the region, while another four thought there were more than 500. There was considerable uncertainty regarding this question, with four panelists responding ‘no opinion/don’t know.’ This implies that the estimation of bobtails is difficult, although anecdotal evidence suggests estimates on the higher side are appropriate; one railroad representative stated that his firm probably generates a few hundred bobtail trips per day.
Source: CS analysis of C-TIP Delphi responses.
There was slightly more certitude about the share of bobtails that could be eliminated with C-TIP. Table 4.5 shows that most of the panelists (9 out of 15) felt there was scope for at least a 16 percent reduction in total daily bobtail trips in Chicago through a C-TIP deployment. Respondents noted that lack of visibility on the other end of cross-town drays creates missed opportunities for matching loads, and that C-TIP’s visibility, notification, and prediction capabilities could mitigate this.
Source: CS analysis of C-TIP Delphi responses.
Idle Time
There was general agreement in the first Delphi round about the average time trucks spend idling in the terminal per trip; in fact, follow up was not required for these questions because consensus was achieved in the first round. As shown in Table 4.6, all 15 panelists felt average idle times in Chicago were 60 minutes or less, and a plurality (8 out of 15) thought they were 30 minutes or under.
Source: CS analysis of C-TIP Delphi responses.
Given the general agreement that idle time in the terminal already is minimal, panelists only estimated mild idle reduction from a theoretical Chicago C-TIP deployment (Table 4.7). Eight respondents thought C-TIP would not achieve any reduction in terminal idle times if it was deployed in Chicago; five felt it would achieve minimal savings of up to 15 minutes per trip on average, while one panelist estimated idle reduction of 16-30 minutes per trip.
The general consensus appears to be that railroads already manage dwell time quite well. Waiting time outside the terminal gates is another matter, but panelists did not perceive C-TIP to offer much benefit in that regard. However, reductions in unproductive bobtail trips may reduce queues outside the gates, and Delphi participants may not have fully considered this.
Source: CS analysis of C-TIP Delphi responses.
Other Questions
The final set of questions dealt with other topics such as key intermodal routes/lanes in the Chicago region; C-TIP benefits that might accrue to the railroads; and C-TIP potential to help rationalize business processes and reduce manual effort in intermodal transportation. Responses to these questions are summarized below.
- Key Chicago cross-town lanes. There are a few high-volume intermodal lanes that were mentioned frequently. Examples include from the BNSF Cicero and Corwith yards to the Norfolk Southern 47th Street and 63rd Street facilities and CSX Bedford Park. Another was from UP Global 2 to the same NS and CSX yards. Overall the prevailing pattern is from western railroads to eastern railroads, which is not unexpected and is similar to Kansas City but greater in scope. These lanes would probably be the most promising if C-TIP technologies were to be deployed in Chicago.
- C-TIP financial benefits for railroads. Panelists generally felt that there may be a benefit to the railroads but it would probably be modest, perhaps in the form of slightly lower rates for cross-town service, or reduced fuel surcharges resulting from fuel savings and the reduction in unproductive trips. This suggests that future deployments of C-TIP technologies should focus on the drayage community, allowing them to find the best operational efficiencies within the markets they serve.
- C-TIP potential to replace manual processes. Most respondents felt that C-TIP could reduce manual processes substantially, though the amount of manual work that might be reduced varied. Savings mostly revolved around track and trace, routing, visibility of bobtailing drivers to dispatchers, and dispatch (having a central dispatch system used by multiple trucking firms). The best opportunities for business efficiencies therefore probably arise from improving the dispatching system for draymen, e.g., by allowing dispatchers to see where drivers are in real-time and dispatch bobtailing truckers who are close to available loads. This is what happened in the drayage optimization tests in Kansas City and Chicago.
