Category

AI Case Study

Using AI to make roller coasters safer

By | AI Case Study, Artificial Intelligence | No Comments

Customer: One of the leading integrated resorts

Summary

The customer an integrated resort on the island in Singapore. They offer some world-class attractions one of which is the Battlestar Galactica, the most enticing duel rollercoaster ride at the resort. They decided to cater to preventive maintenance of the wheels of this ride to ensure top-class safety. They planned to adopt Machine Learning (ML) based solution via Google cloud platform (GCP).

Problem Statement

  • The Customer’s Battlestar Galactica ride is financially quite demanding and requires high maintenance.
  • The wheel detection process is time-consuming and a high maintenance manual job.
  • Decision making on the good versus the bad wheel is based on human judgement and expert’s experience.

The ultimate goal was to remove human intervention and automate the decision making on the identification of a bad wheel using machine learning. The machine learning model needed to be trained on currently available data and ingest real-time data over a period of time to help identify patterns of range and intensity values of wheels. This would in turn help in identifying the wheel as good or bad at the end of every run.

Proposed Solution

Project pre-requisites

Ordering of .dat files generated by SICK cameras to be maintained in a single date-time format for appropriate Radio-frequency identification (RFID) wheel mapping. Bad wheel data should be stored in the same format as a good wheel (.dat files) in order to train the classifier. The dashboard to contain the trend of intensity and height values. Single folder to be maintained for Cam_01 and another folder for Cam_02, folder name or location should not change.

Solution

  • Data ingestion and storage

An image capturing software tool named Ranger Studio was used to absorb the complete information on wheels. The Ranger Studio onsite machine generates .dat files for wheels post every run and stores in a local machine. An upload service picks these .dat files from the storage location at pre-defined intervals and runs C# code on it to provide CSV output with range and intensity values.

CSV files are pushed to Google Cloud Services (GCS) using Google Pub/Sub real-time messaging service. The publisher is used to publish files from the local machine using two separate python scripts for Cam01 and Cam02. The subscriber is then used to subscribe to the published files for Cam01 and Cam02.

  • Data Processing

Powerup is responsible to ingest the data into cloud storage or cloud SQL based on the defined format. Processing of data would include the timestamp and wheel run count. There is a pub tracker and a sub tracker maintained to track the files for both cameras so that the subscribed files can be stored on GCS for both the cameras separately. After CSV data is processed, it is removed from the local machine via a custom script to avoid memory issues.

  • Data modelling Cloud SQL

Once data is processed, Powerup to design the data model in cloud SQL where all the data points will be stored in relational format.

The CSV files of individual wheels are then used to train the classifier model. The classifier model is built with an application programming interface named Keras. The trained classifier helps generate a prediction model (.pkl file) to identify good and bad wheels. The prediction model resides on a GCP VM. The generated CSV files are passed through the prediction model and are classified as good or bad based on an accuracy value.

  • Big Query and ML Model

Once the prediction for a wheel is done, the predicted accuracy score, timestamp and wheel information is stored into the Big Query tables. The average wheel accuracy for wheels is then displayed on Google Data Studio.

Powerup to ensure optimization of data performance via tuning and build the ML model. This would enable the customer to obtain large volumes of height and intensity data, post which, they score the ML model with new data.

Current accuracy threshold for SMS trigger is set at 70. Accuracy of prediction is set to improve over a period 6 months when the model has enough bad wheel data reported for training the ML classifier model. SMS will be triggered if the accuracy value is below 70.

SMS will also be triggered if a file is not received from the local machine to Google Cloud Storage via Google Pub/Sub. The reason for file not being received needs to be checked by the client’s SICK team as it may be due to multiple reasons like source file not generated due to camera malfunction, system shutdown or maintenance and so on. Powerup team to be informed about the same as the restart of instances may be required in such cases. Twilio is the service used for SMS whereas SendGrid is used for email notifications.

  • Security and Deployment

Powerup to build a secure environment for all third party integrations. Deploy User Acceptance Test (UAT) environments, conduct regression tests and provide

Go Live Support to off-site applications. The number of servers and services supported with the production was 10 where support included server management in terms of security, network, DevOps, backup DR and audit. Support also included adjusting ML models to improvise training.

 

Limitations

Since the request payload size was higher, Google ML / Online Predictor could not be used. A custom prediction model was built with Keras to overcome this.

Artificial Intelligence

Cloud platform

Google Cloud Platform.

Technologies used

Cloud Storage, Bog Query, Data Studio, Compute Engine.

Business Benefits

Powerup has successfully been able to train the classifier model with a limited set of good and bad wheel real-time data. The accuracy of the model is expected to improve over time. With current data, the accuracy of the model stands at 60% ensuring cost-effectiveness and world-class safety.

Voice-based personal assistant

By | AI Case Study, Case Study | No Comments

Customer: A global IT & consulting company

About Customer

The customer, which serves clients across six continents has a complex IT landscape to manage. The underlying infrastructure supports a huge employee base and all critical applications, including myApp – the digital platform for self-service that gives employees a seamless experience across various processes and workflows. myApp enables all its employees and contractors to manage business transactions, access productivity tools, news, videos, communications, and other content via one single application interface. Tens of thousands of its employees worldwide depend on MyApp and an associated suite of 150+ applications for their day-to-day activities. But the existing approval-based systems for requests rendered it difficult to handle higher numbers of transactions and larger volumes of data resulting in delays in approvals and decreased employee satisfaction. The customer needed a smart Artificial Intelligence (AI) solution which uses advanced decision-making and machine learning to not only resolve this but also customize the process as per the request while also reducing the number of inputs by the user.

Proposed Solution

Powerup conducted an in-depth study of myApp’s systems and interacted with the users to understand the challenges. The major bottleneck was not the sheer number of requests being received on the portal, but the systems’ inability to understand user context and the number of steps involved in getting simple issues resolved.

Powerup designed a solution for Customers, which will integrate with their myApp portal as a voice engine to automate user journey on the system. This also has to be a voice-first solution that executes an action on voice inputs of the user. The engine backed by strong neural networks understands the user context and personalizes the engine for the user. The engine is built on an unsupervised learning model, where the engine personalizes the conversation based on the user’s past interactions. Thus, providing a unique and easy to navigate through a journey for each user. In this process, the users can get rid of the transactional system and get issues resolved, from approval to task submission, within 2-3 steps.

Powerup also implemented Botzer, chatbot platform with Amazon Lex & Polly. Customer calls get diverted from IVR to the chatbot, which takes customers’ requests as voice input, does entity matching, triggers workflows, and answers back immediately. The voice engine supports 2 languages today – English and Hindi. Customers can get details like Statement of Account, EMI tenure, Balance Due, etc. The intelligence built into the system allows it to behave differently with different users during a different times of the day, thus if the user accesses different applications during the morning than the evening hours, the engine will respond accordingly during the respective hours.

Below is a high-level Solution workflow of the engine, being developed on AWS Lex & Polly, utilizing Botzer APIs at the backend.

 

Following is the high-level technical architecture of the implementation. The engine is hosted on the customer’s AWS VPC, ensuring data integrity & security. The current architecture is capable of hosting 1lakhs+ Customer employee, with 150+ applications on myApp.

Video demo

Technical Architecture

Following is the high-level technical architecture of the implementation. The engine is hosted on the customer’s AWS VPC, ensuring data integrity & security. 

The current architecture is capable of hosting 1lakhs+ employee, with 150+ applications

The Benefit

Faster ticket resolution and better communication with third-party application providers led to an increase in the number of tickets resolved. At the same time, the number of false positives decreased.

Customer support enablement with AWS Connect

By | AI Case Study, AWS, Case Study, Cloud Case Study | No Comments

Customer: A multinational home appliance manufacturer

 

The Problem:

There were several legacy issues with the existing system, as detailed below with the information being provided across categories including service schedules/inquiries, spare part status, service location for maintenance, product information, etc.

  • No Call Recording facility from Avaya
  • No Historical Data and Reports generation. Agents were manually generating reports daily and then aggregating them on excel every week for the weekly report
  • Public Holiday Announcement & Operational Hours changes – Ex: During Ramzan, it’s closes early, involved doing a manual recording and deploying it on the server
  • Scalability issues: A limit of 12 in a queue based on the support from the existing systems – 8 for inbound calls and 4 outbound and concurrent inbound calls
  • Average speed of answering calls was 35 seconds

The approach:

The client wanted to do a pilot project using Amazon Connect, moving from their current voice system hosted in their Mumbai region to Amazon services to achieve the following functionalities:

  1. ‎Ability to take voice calls
  2. On-call connect, an option to choose a language (English/Bahasa)
  3. Call routing based on the language proficiency of the agent
  4. Ability to record calls
  5. Ability to help supervision of calls
  6. Ability to transfer/conference calls
  7. Scalable environment
  8. The ability to generate records in real-time

Solution flow & design:

 

The steps:

  1. Customer calls to the service center number
  2. The Call is routed to AWS Connect through Twilio or equivalent ISP
  3. As per the routing profile, AWS Connect directs the call to the agent
  4. Agent will get a notification in Instaedge CRM of the incoming call, if the mobile of incoming matches with any record in the customer Database, the customer information will be displayed in the Instaedge
  5. The agent will have to log into the Connect panel separately with credentials.

AI-based solution

By | AI Case Study, Case Study, Cloud Case Study | No Comments

Customer: A leading food and agri-business company in the world

Problem Statement

One of our client is a leading food and agri-business company in the world was in the process of building an E-Commerce application for their products to ensure global access & availability. They were in need of a solution which gives them complete visibility into their Micro-services & PAAS architecture and track all the application transaction rather than a sampling of them. They also wanted visibility in User Analytics so they can analyse the conversion trends & user behaviour in the context of User Session.

With above complexity they needed a solution that gives them Automated Problem Detection & Root Cause analytics so they can focus on the findings and make the end-user experience smoother rather than investing time in finding the root cause of those problems.

Proposed Solution

After thorough evaluation Powerup recommended the use of an AI-based solution which can automatically analyse all the dependency at micro-services level and can also trace the Root Cause at code-level depth. For this Powerup leverage the capabilities & offerings of Dynatrace APM tool.

The approach

Implementation stage: Powerup implemented Dynatrace by deploying a one agent on the Kubernetes host which initiated monitoring of all the Micro-services. Within a few minutes Dynatrace could automatically discover the Application Topology map with dependencies.

Powerup also integrated Azure PAAS Service with Dynatrace to gain complete visibility in application.

Configuration stage

  1. Management zone: Powerup configured different management zones so different teams can have the visibility of relevant data.
  2. User Tagging: Powerup configured user session tagging, Key User actions and set up the conversion goals to track the revenue over user experience.
  3. Dashboard: Powerup created the all in one dashboard so in single view they can track the User Experience, Application Transactions status, Infrastructure health, API Calls and Problem detection.

Dynatrace applied Dynamic thresholding on all the detected anomalies, Powerup helped customer to understand and analyse the automated detected problems and trace the Root Cause.
Powerup ensures High availability & quick content delivery of application at a global level by managing the PAAS services in HA mode & CDN to ensure quick response.

Cloud platform

AZURE

Technologies used

Dynatrace One Agent, Dynatrace DEM, Kubernetes, AZURE PAAS Services, CDN

How we helped a leading BFSI corporation improve process efficiency by 60% using AI-based OCR.

By | AI Case Study, AWS, Case Study | No Comments

Written by Vinit Balani, Associate product manager at Powerupcloud Technologies

 

Demonetization has changed the way the Indian banking sector functions. While the wider acceptance of Aadhaar has made documentation and authentication easier, for BFSI clients mainly the insurance companies, document verification is still required for processing of loans and policies. Most of this process still remains manual, adding to the time required for opening an account or processing the claims.

One of India’s largest insurance companies was facing this challenge and wanted to resolve this problem. In this case study, we highlight the problem statement and take you through how Powerupcloud came up with the solution using AI automation.

 

Problem Statement:

With a customer base of 115 million users and expanding, one of India’s largest private Insurance companies wanted a resolve the problem where field agents were dealing with a large quantum of information including images captured as a part of a KYC process.

The company already had an android application for these agents to capture photos of documents, while an account was being created. However, once these photos were captured the data had to be manually entered into the customer repository.

Another challenge was the bad quality of document photographs being taken by the agents, which often resulted in them going back to the customer to re-capture the photos. This was ultimately also increased lead-time for account opening.

The company was looking to automate this process using an OCR (Optical Character Recognition) solution, which could help approve or reject the photo based on quality at the point of capture itself.

 

Proposed Solution:

Powerup proposed creating a native android application to be integrated within the company’s primary application by leveraging AWS Rekognition’s OCR technology. In addition an application with features to improve the image and do a quality check using open source technologies.

The current scope included developing an OCR mechanism for only the Aadhaar Card. With its successful implementation, it is now going to be extended to other KYC documents.

 

Solution Flow:

 

 

Solution Details

The app (OCR) developed by Powerup is a native android app integrated within the company’s existing android app. The OCR app gets triggered when the Aadhaar document has to be captured by the field agent.

Once the image is captured, it allows the user to crop and enhance the image using features like brightness, contrast, saturation, etc. Post this, the image quality check is done for brightness, contrast, and blurriness. If the image fails the quality check, the agent is asked to re-capture it. However, if it passes the test, the Aadhaar image is scanned to check for QR code and extract information from it. If the scan succeeds, the output with parameter values (like Aadhaar number, Name, Gender, Address) is sent to the company’s application. However, if the QR scan is not successful, the text/parameters are extracted from the image using AWS Rekognition’s OCR technology.

The extracted parameters are then passed on to the company’s Android application as JSON. Below are some snapshots of the native OCR app –

 

 

 

Cloud platform

AWS.

Technologies/Features/Services used

AWS Rekognition, Python.

Benefit

The application is now live and being used by 6000+ field agents across India. There has led to a 60% reduction in the lead-time for processing an application. In addition, the solution has also helped improve the productivity of the field agents who can now cover a lot more customers.

Automated photo moderation

By | AI Case Study, AWS, Case Study | No Comments

Customer: Shaadi.com

A leading matrimony site in India

Problem Statement

A leading matrimony site in India receives 20,000 new profile creations every day.
A team of 16 reviews the uploaded profile pictures and moderates them based on
9 parameters including nudity, celebrity, blur, group photos, photoshopped images, etc. The customer wanted to automate this moderation process to improve efficiency and reduce manpower costs.

Proposed Solution

Powerup used a combination of Amazon Rekognition and custom rule engine to moderate the images in real-time. The solution was consistently achieving above
80% accuracy. This brought down the moderation time from 24 hours to 3 minutes and the headcount was reduced from 16 to 4.

Cloud Platform

AWS.

Technologies used

Amazon Rekognition, Lambda, OpenCV, Python.

App Voice Automation

By | AI Case Study, AWS, Case Study | No Comments

One of the largest Telcos in India

Problem Statement

one of the largest Telcos in India has developed an app that allows its subscribers
to query their account details & get regular updates on the same. Currently, 2
million+ people use the mobile app. They wanted to provide a better user interface,
which allows users to get better results, with minimum text.

Proposed Solution

Powerup integrated with AWS Lex & Polly to provide complete text-based
automation on the App. The users are allowed to search for latest offers, their
consumption & bill payment with voice commands, thus allowing them to get the latest
updates with minimal text intervention.

Cloud Platform

AWS.

Technologies used

Botzer, Lex, Polly.

Virtual Receptionist Bot

By | AI Case Study, Azure, Botzer, Case Study, Chatbot | No Comments

One of the world’s largest IT park builders

Problem Statement

One of the world’s largest IT park builders was looking to automate their front-desk
assistant roles to cater to a large volume of visitors, better response and added
security.

Proposed Solution

Powerup implemented botzer.io – a receptionist bot which takes the image of the
visitor, checks the database for repeating/new visitor, checks the calendar for
appointment and prints visitor badge with access code. The visitors can also interact
with the bot through chat or voice asking for directions within the building. Botzer
A receptionist is integrated with Google Nearby to answer queries like nearby taxi
stand, metro station, coffee shops, restaurants, etc.

Cloud Platform

Microsoft Azure.

Technologies used

Botzer.io, Azure Face API.

Email Classification Engine

By | AI Case Study, Case Study, ML | No Comments

Indian financial services company

Problem Statement

An Indian financial services firm gets close to 1.2 million support emails per month.
They outsourced email categorization to a 50-member support team. Each support
member had to read 2.27 emails per minute and classify them in the right bucket.
Customer faced several issues including 24-hour SLA, team attrition and the
classification accuracy was less than 80%.

Proposed Solution

Powerup built an email classification engine using a combination of several
algorithms & techniques including a bag of words, stemming, lemmatization, decision
forest, neural networks and more. The machine learning was able to achieve above
80% accuracy on a consistent basis.

Workflow

Technologies used

Spark, Python, SQL Server.

POS Automation

By | AI Case Study, AWS, Case Study, Data Case Study | No Comments

Joint venture of world’s largest banking and financial services groups

Problem Statement

The customer has a user base of 6 million across India & have a major presence in tier 2
& 3 cities. They have a sales team of more than 8000 field agents that visit
customers for background verification, new policy registrations, etc. Most of their
tasks involve capturing user photo ID details for verification/registration. This
happens either via manual entry or by clicking an image of the photo ID.

Proposed Solution

Powerup developed a POS automation, as an Android application, which allows
sales agents to capture real-time images, performs image pre-processing to improve
the image quality & extracts data real-time for their perusal & sends the data to
backend system for user storage.
The solution helped increase sales team efficiency by more than 15%.

Cloud Platform

AWS.

Technologies used

AWS Rekognition, Python.