Large enterprises that provide services to consumers may receive millions of customer complaint tickets every month. Handling these tickets on time is very critical, as this directly impacts the quality of service and network efficiency.

A ticket may be assigned to multiple teams before it gets resolved. Assigning a ticket to an appropriate group is usually done manually as the complaint information provided by the customer is not very specific and maybe inaccurate sometimes. This manual process incurs enormous labor costs and is very time inefficient as each ticket may end up in the queue for hours.

In this talk, we will present an approach to automate the process of ticket routing completely. We will start by discussing how we can use Markov Chains to model the flow of tickets across different teams. Next, we will discuss the feature engineering part and why Factorization Machine Models are essential for such a use case. This will be followed by a discussion on the learning of decision rule sets in a supervised manner. These decision rules can be used to traverse tickets across multiple teams in an automated fashion. Thus, automating the complete process of ticket routing. We will also discuss that the proposed framework can be validated easily by SMEs, unlike other AI solutions, thus, resulting in its quick acceptability in an organization. Finally, we will go through the different settings in which this solution can fit, therefore, resulting in its broad applicability.

The framework can provide substantial cost savings to enterprises. It can also reduce Response time to tickets significantly by almost eliminating the queue time. Overall, it can help large enterprises in

1. Saving costs by reducing the workforce of ticket handling team

2. Increasing revenue by improving quality of customer experience

 
 

Outline/Structure of the Talk

1. Problem Definition & Motivation – 5 Mins

1.1 An overview of ticketing system

1.2 Journey of a ticket

1.3 Scale of Problem

2. Details of Proposed Framework – 12 Mins

2.1 Modelling of ticket flow using Markov Chains

2.2 Feature Engineering

2.2.1 Factorization Machine Models

2.3 Decision Rules Generation

2.4. Advantages of Framework

3. Questions – 3 Mins

Learning Outcome

1. How ticket routing problem can be framed as a Data Science problem.

2. Techniques to be used while automating ticket routing.

3. Challenges faced while deploying automated ticket routing to production.

Target Audience

Applied Data scientists, CXOs from Consumer service providers, Operations Leaders

Prerequisites for Attendees

Basic understanding of Machine Learning

schedule Submitted 5 months ago

Public Feedback

comment Suggest improvements to the Author
  • Deepti Tomar
    By Deepti Tomar  ~  4 months ago
    reply Reply

    Hello Priyanshu,

    Thanks for your proposal! Since we are considering 20 mins proposals on the submission system, request to update the outline/structure with a time break - up for 20 min with 1 speaker.

    Thanks,

    Deepti

    • Priyanshu Jain
      By Priyanshu Jain  ~  3 months ago
      reply Reply

      Hi Deepti,

      I have updated the structure of proposal to 20 mins. 

       

  • Ashay Tamhane
    By Ashay Tamhane  ~  4 months ago
    reply Reply

    Hi Priyanshu, thanks for an interesting proposal. Could you clarify if there are other techniques that you compared this approach against? How did this approach fare against those baselines? Thanks.

    • Priyanshu Jain
      By Priyanshu Jain  ~  4 months ago
      reply Reply

      Hi Ashay,

      Thanks for reviewing my proposal.

      Yes, I have compared this approach against others.

      I have updated the document under Links section. Please refer to Section. 2 of the shared document for detailed comparison.

      Let me know if anything else is required from my end.

      Regards,

      Priyanshu

      • Ashay Tamhane
        By Ashay Tamhane  ~  4 months ago
        reply Reply

        Hi Priyanshu, thanks for the detailed doc. Is accuracy is a good metric for you - does it lead to any bias due to class imbalance (if any)? Are there any other metrics that you looked at?

        • Priyanshu Jain
          By Priyanshu Jain  ~  3 months ago
          reply Reply

          Hi Ashay,

          In our experiments, there were very few cases with high class imbalance. Hence, we continued with accuracy as it is easy to explain to stakeholders. However, we did check AUC-ROC curve for high class imbalance cases and the results were inline with the results shared in the document.

           

  • Sujoy Roychowdhury
    By Sujoy Roychowdhury  ~  5 months ago
    reply Reply

    1. How is the blog link relevant to the talk ?

    2.  will you be comparing the results of this method with other methods ? 

    • Priyanshu Jain
      By Priyanshu Jain  ~  5 months ago
      reply Reply

      Hi Sujoy,

      The blog is not related to the talk. I think I didn't fully understand what could be shared as links.

      Yes, I will be comparing the proposed method with other methods. The comparison will mostly be around the applicability of methods in production setting.

       

      • Sujoy Roychowdhury
        By Sujoy Roychowdhury  ~  5 months ago
        reply Reply

        Any writeup / publication / slides which will help us better understand how you have used Markov chains in ticket routing.

        • Priyanshu Jain
          By Priyanshu Jain  ~  5 months ago
          reply Reply

          Hi Sujoy,

          I have added a (work in progress) write-up on the complete approach under Links section. I hope this helps.

          Let me know if anything else is required.

          Regards,

          Priyanshu

  • Deepti Tomar
    By Deepti Tomar  ~  5 months ago
    reply Reply

    Hello Priyanshu,

    Thanks for your time and efforts on the proposal! This is a well structured and detailed proposal.

    Could you answer the following questions to help the program committee understand your proposal better?

    • Are these demo(s)/use case(s) from your project work (industry-specific use cases)? Speaker's experience on the project helps people understand the concept better.
    •  It's great to see that you've mentioned in learning outcomes about the Challenges faced in the implementation of the technique in the application. Adding this to your outline/structure along with the workarounds would be helpful.
    • Has the solution been implemented already and results ( saved cost & increased revenue ) available to be shared with attendees?

    Thanks,

    Deepti

    • Priyanshu Jain
      By Priyanshu Jain  ~  5 months ago
      reply Reply

      Hi Deepti,

      Thanks for reviewing my proposal.

      Please find below the answers to your questions - 

      1. Yes, this proposal is based on a project I did recently at Guavus.

      2. Sure. I am planning to discuss this as a part of '2.4. Advantages of Framework'

      3. Yes, I have some results which can be shared with attendees.

      Let me know if any other information is required from my end.

      Regards,

      Priyanshu

      • Deepti Tomar
        By Deepti Tomar  ~  5 months ago
        reply Reply

        Great, Thanks for your response Priyanshu! We will let you know in case if we have more questions.

  • Natasha Rodrigues
    By Natasha Rodrigues  ~  5 months ago
    reply Reply

    Hi Priyanshu,

    Thanks for your proposal!

    To help the program committee understand your presentation style, can you provide a link to your past recording or record a small 1-2 mins trailer of your talk and share the link to the same?

    Also, in order to ensure the completeness of your proposal, we suggest you go through the review process requirements.

    Thanks,
    Natasha

    • Priyanshu Jain
      By Priyanshu Jain  ~  5 months ago
      reply Reply

      Hi Natasha,

      I have uploaded a short video of a demo which I gave during hackathon in my office. Hope this helps.

      I have also shared link of a blog which I wrote for Guavus sometime back.

      Let me know if anything else is required from my end.

      Regards,

      Priyanshu

      • Natasha Rodrigues
        By Natasha Rodrigues  ~  5 months ago
        reply Reply

        Hi Priyanshu,

        Many thanks for the updates.

        Regards,

        Natasha