Image Alt

Mobility Innovators

How ChatGPT will be game changer for Public Transit agencies?

ChatGPT is the latest buzzword as everyone is trying to figure out how this tool will impact different industries. If you are not aware of what is ChatGPT? It is a chatbot launched by OpenAI in November 2022, built on top of OpenAI’s GPT-3 family of large language models. It is a generative tool that can get you answers within a few seconds.

Public Transport or Mobility is a traditional sector so we always feel that this will not impact the sector. Many people feel that ChatGPT has no relevance for the public transit sector. They may be true as ChatGPT is just compiling information from different sources and providing it in a synchronized way. This may be true but may be true for ChatGPT 1.0. These can be very powerful tools in the coming years.

Why not ask ChatGPT what it can do for the Public Transit sector:

Q1. How ChatGPT will help the public transit sector?

ChatGPT can help the public transit sector in a number of ways, including:

  1. Automating customer service: ChatGPT can be used to build conversational interfaces that can answer common questions from customers, reducing the need for human customer service representatives.
  2. Providing real-time information: ChatGPT can be integrated with data sources such as transit schedules and real-time location information to provide customers with accurate, up-to-date information about their trip.
  3. Personalizing the experience: ChatGPT can be used to build personalized experiences for customers, such as providing customized route recommendations based on their preferences and past travel history.
  4. Improving efficiency: ChatGPT can be used to automate various back-office tasks, such as schedule optimization and workforce management, helping to improve the overall efficiency of the public transit system.
  5. Improving safety: ChatGPT can be used to monitor the real-time location of the vehicles and can alert the driver if there is any emergency, this will help to improve the safety of the public transit system.

Q2. How ChatGPT will impact the public transit sector?

ChatGPT, as a language model, can potentially be used in the public transit sector to improve customer service through natural language understanding and generation. It could be used in chatbots for answering passenger inquiries, providing real-time information about transit schedules and routes, and even assist with ticket purchasing. Additionally, it can be used to analyze large amounts of data generated by the transit system to identify patterns and improve overall efficiency. However, it’s important to note that ChatGPT is a tool, the actual impact will depend on how it is implemented and integrated into the specific use case.

Q3. How it can impact the work of public transport consultant work?

ChatGPT, as a language model, can potentially impact the work of a public transport consultant in several ways:

  1. Data analysis: ChatGPT can be used to analyze large amounts of data generated by the transit system, such as passenger counts, travel times, and route usage. This can help consultants identify patterns and trends that can be used to improve the overall efficiency of the transit system.
  2. Simulation and modeling: ChatGPT can be used to simulate and model different scenarios and configurations of the transit system. This can help consultants evaluate the potential impact of proposed changes and identify potential issues before they occur.
  3. Report generation: ChatGPT can be used to generate reports and presentations that summarize the findings of data analysis and simulation. This can help consultants communicate their findings to clients and stakeholders in a clear and concise manner.
  4. Chatbot for customer service: ChatGPT can be integrated into a chatbot for answering passenger inquiries, providing real-time information about transit schedules and routes, and assist with ticket purchasing. This can help consultants to improve the customer service provided by the transit system.

It’s important to note that ChatGPT is a tool and its impact will depend on how it is implemented and integrated into the specific use case.

Q4. Can it help in research work too?

Yes, ChatGPT can potentially help in research work in the public transit sector. Here are a few examples:

  1. Natural Language Processing (NLP): ChatGPT can be used to analyze and understand large amounts of unstructured data, such as passenger feedback, social media posts, and news articles. This can help researchers identify patterns, trends, and insights that can be used to improve the transit system.
  2. Text generation: ChatGPT can be used to generate realistic, human-like text, such as passenger feedback, social media posts, and news articles. This can help researchers to understand how people talk about the public transit system, and the concerns they might have.
  3. Survey generation: ChatGPT can be used to generate survey questions or feedback forms to be sent to the public, this can help researchers to gather a large amount of data in a short period of time and understand the public perception of the transit system.
  4. Literature review: ChatGPT can be used to assist researchers in the process of literature review, by summarizing large amount of documents and papers, this can help researchers to have an overview of the state of the art in a certain field, and to identify gaps in the existing research.

It’s important to note that ChatGPT is a tool and its impact will depend on how it is implemented and integrated into the specific use case.

Q5. Can you prepare survey questions for passenger feedback and service improvement?

Sure, here are some examples of survey questions that could be used to gather passenger feedback and gather ideas for improving public transit service:

  1. Overall, how satisfied are you with the public transit service in your area?
  2. How often do you use public transit?
  3. How would you rate the punctuality and reliability of the public transit service?
  4. How would you rate the cleanliness and maintenance of the public transit vehicles and stations?
  5. How would you rate the helpfulness and friendliness of the public transit staff?
  6. Are there any specific routes or times of day where you have experienced crowding or delays?
  7. Are there any specific routes or services that you would like to see added or improved?
  8. Are there any accessibility issues that you have encountered while using public transit?
  9. How would you rate the overall value for the cost of the public transit service?
  10. Are there any other comments or suggestions that you would like to share about the public transit service?

It is important to note that the survey questions should be tailored to the specific use case and the specific public transit system being studied, and the survey should be tested for usability and comprehension before being distributed to the public.

Q6. Can you share an evaluation matrix to check survey responses?

An evaluation matrix can be used to analyze survey responses and help identify areas for improvement in public transit service. Here’s an example of a matrix that could be used:

  1. Overall satisfaction: Responses to the question “Overall, how satisfied are you with the public transit service in your area?” are rated on a scale of 1-5, with 1 being very unsatisfied and 5 being very satisfied.
  2. Frequency of use: Responses to the question “How often do you use public transit?” are grouped into categories such as daily, weekly, monthly, and rarely/never.
  3. Punctuality and reliability: Responses to the question “How would you rate the punctuality and reliability of the public transit service?” are rated on a scale of 1-5, with 1 being very poor and 5 being very good.
  4. Cleanliness and maintenance: Responses to the question “How would you rate the cleanliness and maintenance of the public transit vehicles and stations?” are rated on a scale of 1-5, with 1 being very poor and 5 being very good.
  5. Staff helpfulness and friendliness: Responses to the question “How would you rate the helpfulness and friendliness of the public transit staff?” are rated on a scale of 1-5, with 1 being very poor and 5 being very good.
  6. Crowding and delays: Responses to the question “Are there any specific routes or times of day where you have experienced crowding or delays?” are grouped into categories such as morning commute, evening commute, weekends, and no issues reported.
  7. Additional routes and services: Responses to the question “Are there any specific routes or services that you would like to see added or improved?” are grouped into categories such as new routes, route extensions, additional buses/trains, and other suggestions.
  8. Accessibility issues: Responses to the question “Are there any accessibility issues that you have encountered while using public transit?” are grouped into categories such as wheelchair accessibility, visually impaired accessibility, hearing impaired accessibility, and no issues reported.
  9. Value for cost: Responses to the question “How would you rate the overall value for the cost of the public transit service?” are rated on a scale of 1-5, with 1 being very poor value and 5 being very good value.
  10. Additional comments and suggestions: Responses to the question “Are there any other comments or suggestions that you would like to share about the public transit service?” are grouped into categories such as positive feedback, negative feedback, and suggestions for improvement.

It is important to note that this is just one example of an evaluation matrix and the matrix should be tailored to the specific use case and the specific public transit system being studied.

Q7. Can you evaluate riders’ experience based on Transport for London (TfL)’s twitter feed?

Evaluating riders experience based on Transport for London’s Twitter feed would involve analyzing the tweets posted by TfL and the responses to those tweets. It would also require identifying patterns and trends in the data, and comparing it to riders feedback from other sources such as surveys or interviews.

Here are a few examples of how Transport for London’s Twitter feed could be analyzed to evaluate riders experience:

  1. Sentiment analysis: By performing a sentiment analysis on the tweets and responses, it would be possible to determine the overall tone of the conversations and identify any recurring themes or issues that riders are experiencing.
  2. Volume of tweets: By analyzing the volume of tweets and responses, it would be possible to identify any spikes in activity which could indicate particularly high levels of dissatisfaction or a specific issue impacting riders.
  3. Keyword analysis: By analyzing the keywords used in tweets and responses, it would be possible to identify any specific issues or concerns that riders are raising.
  4. Responding to complaints: By analyzing the way Transport for London is responding to complaints, it would be possible to evaluate the level of customer service that riders are receiving.

It is important to note that this kind of analysis would be a sample of riders experience and it would be better to have a wider feedback mechanism to get a more comprehensive understanding of riders experience.

Q8. Do sentiment analysis using TfL Twitters Feed (https://twitter.com/TfL)?

As a language model AI, I am not able to perform sentiment analysis on a live twitter feed like https://twitter.com/TfL. However, I can provide you with an overview of how sentiment analysis could be performed on Transport for London’s Twitter feed:

  1. Data collection: The first step would be to collect the tweets and responses posted on the TfL Twitter feed. This could be done using a programming library or software specifically designed for collecting data from Twitter, such as the Twitter API.
  2. Data pre-processing: The collected data would then need to be cleaned and pre-processed. This would involve removing any irrelevant information, such as URLs or hashtags, and standardizing the text.
  3. Sentiment analysis: Once the data is cleaned and pre-processed, a sentiment analysis algorithm could be applied to it. There are various techniques that can be used for this, such as natural language processing (NLP) or machine learning (ML) algorithms. These techniques will help in determining the overall sentiment of the tweets and responses, such as positive, negative, or neutral.
  4. Results interpretation: The output of the sentiment analysis would be a sentiment score for each tweet and response, which could then be analyzed to identify any patterns or trends in the data. For example, a higher number of negative sentiment tweets could indicate that riders are experiencing dissatisfaction with a particular aspect of the service.

It’s important to note that sentiment analysis is not always 100% accurate. The context and the way the words are used, can affect the sentiment analysis outcome.

Q9. Can you write code to conduct sentiment analysis for Transport for London (TfL) twitters feed?

Conducting sentiment analysis on social media data, such as tweets, can be a complex task that requires a combination of natural language processing (NLP) techniques and machine learning (ML) algorithms. Here is an example of Python code that uses the tweepy library and the TextBlob library to conduct sentiment analysis on tweets from Transport for London’s (TfL) Twitter feed:

This code uses the tweepy library to authenticate with the Twitter API and retrieve the latest tweets from TfL’s Twitter feed. Then, it uses the TextBlob library to conduct sentiment analysis on the tweets. The TextBlob library returns a sentiment object, which has properties for the polarity and subjectivity of the text. The polarity score is a float within the range [-1.0, 1.0] where -1.0 is very negative, 1.0 is very positive and 0.0 is neutral.

It’s important to note that this is just an example and it is a basic one, it will not work as is since you need to replace the placeholders with your own credentials. Additionally, this is a general approach, the accuracy of the results may vary depending on the complexity of the language used in the tweets, the context of the tweets, or even the subjectivity of the analysis. You may need to fine-tune the sentiment analysis or use more sophisticated NLP techniques to get more accurate results.

Q10. Can you share the code to integrate ChatGPT with Transport for London (TfL) website to provide real-time information about transit schedules and routes in London?

Integrating ChatGPT with a website, such as the Transport for London (TfL) website, to provide real-time information about transit schedules and routes in London would likely involve several steps, including building a web-based interface to interact with ChatGPT, connecting to TfL’s API to retrieve the transit data, and using natural language processing (NLP) techniques to understand and respond to user queries.

Here’s an example of Python code that demonstrates how you might use the OpenAI GPT-3 API to interact with ChatGPT and retrieve transit information:

Can we say that these answers are perfect or accurate? Maybe not but this can make it the consultant, developers, and experts so easy. M&M is always the best, i.e. Man and Machine. 

The content is generated using ChatGPT and the Image is generated using DALL.E.2

About The Author

Jaspal Singh is the Founder of Mobility Innovation Lab (MIL) and Host of the Mobility Innovators Podcast. If you are working on innovative ideas and solving mobility and transportation issues, please feel free to reach out. He loves to talk about startups, mobility, and technology.

Comments

  • JSingh
    Manoj Aggarwal
    February 3, 2023

    Wow. I am sure more and more people will figure out ways to work smarter and quicker using tools such as these. Thanks for posting.

  • JSingh
    Maghsoud Saghaimaroof
    February 24, 2023

    How will ChatGPT affect HOT High Occupancy Toll systems on Freeways?

  • JSingh
    Maghsoud Saghaimaroof
    February 24, 2023

    ChatGPT might help us determine a much smarter Transaction Generation for HOT Users. It might advance our Back Office Processing.

Sorry, the comment form is closed at this time.