Hi There, I'm Pranshu Tiwari

Data Science and Strategy

Seasoned Data Scientist, AI, and Innovation Consultant with 18+ years of expertise driving AI and digital transformation in the energy, insurance, and engineering sectors. Experienced in Machine Learning, Natural Language Processing, and advanced Mathematics to develop predictive models and optimize processes.An award-winning innovator and researcher with thought leadership skills, have filed 30 patents at IBM and Schneider Electric. Pransh has a double Master’s in Data Science and Mathematics-Operations Research from the University of Wisconsin and Northeastern University Boston.

Pranshu Tiwari

About Me

Address

46, Meadow North Grafton, MA 01536

Email

pranshu.ti12@gmail.com

Phone Number

(612) 716-3694

LinkedIn

linkedin.com/in/pranshu-tiwari-18973a10/

Pranshu Tiwari has 18+ years of expertise as a Data Scientist, AI, and Innovation Consultant driving AI and digital transformation in the energy, insurance, and engineering sectors.. He has MS – Operations Research from Northeastern University-Boston and is a project away from completing his third master’s in data science from University of Wisconsin. He has done MBA from Paris Tech -India and B.E in Electrical Engineering from Madras University. His data science Projects include FOAK Drug Recommendation System, Detect Cancer from CT scans of Brain for academia as well as Text Processing based Supervised Machine Learning. He has leadership experience of 10+ years in field of Consulting & Program Delivery and 4 Granted Patents with 14 Filed Patents in IBM. With Depth and Breadth of deep educational and industrial experience across multiple geographies, he is poised to take challenging opportunities to help companies navigate through various Industrial imperatives of AI/Digital and Cloud. He is avid Tennis player and competes in USTA 4.0 and regional tournaments. He also has interest exploring various Meditation techniques personal and spiritual growth

Tennis

Pranshu is an active tennis player. He has a serve and volley game and has a single handed backhand. He plays singles /doubles twice a week. He competes at USTA, Boston Tennis League. You can refer his latest tennis record on the following link:

Link 1

Link 2

Education

2002
University of Madras, India

B.E. in Electrical Engineering

2005
Ecole Ponts Business School, GBS India

ENPC-MBA

August 2021
Northeastern University, Boston

Master of Science, Mathematics Operations Research

2021
University of Wisconsin, Eau-Claire

Master of Science, Data science

Highlights - Digital & AI

Artificial Intelligence

  • Developed FOAK Persona based Drug Recommendation System based on Content Filtering & Comparison of Predicted Rating based on Different Pre-Processing Architectures
  • 90% performance in Brain Cancer Detection using CNN Filters
  • 80% Accuracy in Predicting Stock Price Direction based on Different Pre-processing Architectures and RNN models
  • Created First of a Kind IBM patented AI based Dev Ops Branching Technique which minimizes build prediction failures
  • Developed FOAK IBM Patented Electrical Device Degradation System
Digital / Technology Strategy
 
  • Developed FOAK MDM based Revenue Assurance System for Client in India
  • Won Mega Deal for Revenue Assurance Deal in Mexico
  • Created a Data Quality Roadmap for Utility in UK to improve its data quality
  • Performance Test Strategy for Leading Utility in US
  • Led a team for Successful Digital Re-invention Program to compare and Contrast for advanced Digital Programs
  • Mobile Commerce Strategy for Leading Telco in Malaysia
  • Enterprise Data Quality Roadmap for a Utility in UK
  • Customer Transformation Roadmap for Improved Experience in US
  • Unique ID Report Assessment to Govt of India

Key Skills

Artificial Intelligence

 

  • Bayesian Theory & Maximum Likelihood
  • Unsupervised Machine Learning
  • Gradient Descent & Supervised ML
  • Matrix Vector Decomposition
  • Artificial Neural Network
  • Convolution Neural Network
  • Fourier & Gabor Transformation
  • Natural Language Processing
  • Linear & Integer Programming
  • Markov Chain & Genetic Algorithm
  • Tensor Flow, Kera’s, NumPy, Sklearn & R Libraries

Technologies

 

  • Programming Languages: Python, R,SQL
  • Libraries: NumPy, SciPy, OpenCV, Scikit, Keres, Glove, PyTorch,Caret
  • Machine Learning and Deep Learning Frameworks: Convolution, LSTM,
    Homography, Feed-Forward
  • Generative Models: Auto-Encoder, LLM, Topic Modelling
  • Version Control and Collaboration Tools: GIT
  • Process Mapping: Visio Process Maps
  • Neural Networks & Deep Learning Frameworks: NNET, Keres, RESNET, VGGNet,
    PyToch
Digital / Technology Strategy
  • Client Experience Management
  • Application Rationalization
  • Transformation Roadmap
  • Cloud Migration Assessment
  • Information Strategy Roadmap
  • Business Analysis
  • Mobile Commerce Strategy
  • Project Management-Agile & SDP
  • Master Data Management Strategy
  • RFP & Vendor Selection

PORTFOLIO - PRANSHU TIWARI

download (1) (1)

Insurance Risk Rating based on Maximum Lkelihood Estimation

Tweeter analysis

Tweeter Analysis to cmpare Branding of 2 Sports Companies

How health care

Insights from Patient Data for Disease Predction & Dynamic pricing 

Markov

Markov Based Models for Time Series Predicion of Energy States

Persona based

Persona based Drug Recommendation Systm based on Machine Learning

newspapers

RNN and  Likelihood based Models for Stock Price Prediction based on Daily News

detect-small

Model Evaluation on Image Data Set- Faster RCNN vs Yolo V5

predictive-warehouse-management-1024x676

Forecasting Supply Chain

PROFESSIONAL EXPERIENCE

2005 - 2006
Research Analyst
I Bilt
2006 - 2008
Associate Consultant
Wipro
2009 - 2011
Lead Consultant
IBM
2011 - 2014
Advisory Consultant
IBM
2014 - 2017
Senior Advisory Consultant
IBM
2017 - 2021
Managing Consultant
IBM

Data Scientist | IBM Senior Inventor

2021 - 2023
AI Data Scientist & IBM Master Inventor
IBM

AI Data Scientist & IBM Master Inventor

2023- 2025
AI Consultant
AI Consultant

Schneider Electric | Principal AI Consultant

Patents and Awards

  • 3 Plateau Award from IBM CEO 2017, 2020

  • Honored IBM Senior Inventor in 2017

  • Manager Choice Award 2014, 2017

  • Treasured Wild Ducks 2016

  • Shared Expertise 2016

  • Top Contributor 2015, 2016,2017

Certifcates

Carnegie Mellon University - Computer Vision

University of California-San Diago- Machine Learning Methods

PATENTS BY INVENTOR PRANSHU TIWARI

Pranshu Tiwari has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

Publication number: 20170017913
Abstract: In an approach to managing data quality and compliance, a computer retrieves data within a business enterprise system. The computer applies a set of rules to the data, wherein the application of the set of rules to the data produces a plurality of mapped data, and the computer aggregates the plurality of mapped data into a plurality of results. The computer then reports the plurality of results for a data display.
 
Type: Application
Filed: July 17, 2015
Publication date: January 19, 2017
Inventors: Harish Bharti, Kshitij K. Raval, Pranshu Tiwari
Publication number: 20230132337
Abstract: A method includes using, by one or more processors of a computer system, a cognitive model to estimate software development parameters for a software development project based on one or more similar past projects, and automatically assigning, by the one or more processors of the computer system, story points to sprints of the software development project based on the estimated software development parameters.
Type: Application
Filed: October 27, 2021
Publication date: April 27, 2023
Inventors: Pranshu Tiwari, Saurabh Trehan, Raghava Venkat, Balakrishnan Sreenivasan, Renny I John, Harish Bharti
Patent number: 12154049
Abstract: A method includes using, by one or more processors of a computer system, a cognitive model to estimate software development parameters for a software development project based on one or more similar past projects, and automatically assigning, by the one or more processors of the computer system, story points to sprints of the software development project based on the estimated software development parameters.
Type: Grant
Filed: October 27, 2021
Date of Patent: November 26, 2024
Assignee: International Business Machines Corporation
Inventors: Pranshu Tiwari, Saurabh Trehan, Raghava Venkat, Balakrishnan Sreenivasan, Renny I John, Harish Bharti

Publication number: 20210241392

Abstract: Methods and systems for metrics for energy saving and response behavior are disclosed. A method includes: receiving, by a computing device, for each of a plurality of energy users, consumption time series data from a smart meter of the energy user; determining, by the computing device, for each of the plurality of energy users, demographic data of the energy user; clustering, by the computing device, the energy users based on the consumption time series data and the demographic data; identifying, by the computing device, a plurality of groups of energy users based upon the clustering; and determining, by the computing device, an energy saving program to associate with each of the plurality of groups.

Type: Application

Filed: February 5, 2020

Publication date: August 5, 2021

Inventors: Pranshu TIWARI, Harish BHARTI, Saurabh TREHAN, Anshul CHETAL

Publication number: 20240220889
Abstract: An end-to-end framework that provides for pair programming can determine a project intent. Programming requirements needed to fulfill the project intent can be identified. The programming requirements can be reduced into lower dimensions. The reduced programming requirements can be clustered into clusters. A common theme can be identified in each of the clusters. From at least one of the clusters having a common theme corresponding to the programming requirements, feasible pairs of developers can be selected. At least one optimal pair of developers can be determined among the feasible pairs using an optimization algorithm that optimizes the project intent.
Type: Application
Filed: December 29, 2022
Publication date: July 4, 2024
Inventors: Pranshu Tiwari, Swarnalata Patel, Saurabh Trehan, Harish Bharti
Publication number: 20240241707
Abstract: In various examples, a computer-implemented method includes: ingesting, by one or more computing devices, application deployment data for an application; generating, by the one or more computing devices, a cloud application deployment predictor data structure for the application; generating, by the one or more computing devices, objective functions for the cloud application deployment predictor data structure for the application; optimizing, by the one or more computing devices, between the objective functions for the application; and generating, by the one or more computing devices, based on the optimizing between the objective functions, a multi-cloud deployment map for the application.
Type: Application
Filed: January 12, 2023
Publication date: July 18, 2024
Inventors: PRANSHU TIWARI, Harish Bharti, Swarnalata Patel, NAVEEN NARAYANASWAMY, Abhaya Kumar Sahoo
Publication number: 20230140410
Abstract: A method includes receiving, by one or more processors of a computer system, historical data related to deployment characteristics and an architecture for past incidents occurring in a data warehouse, predicting, by the one or more processors of the computer system using neural network modeling, tickets related to a response to at least one incident occurring in the data warehouse, wherein the predicting is based on the deployment characteristics and the architecture of the data warehouse, considering, by the one or more processors of a computer system, a plurality of parameters to ascertain the predicted tickets, and providing, by the one or more processors of the computer system, incident ticket volume prediction including the predicted tickets to an incident management system interface reviewable by a user.
Type: Application
Filed: October 28, 2021
Publication date: May 4, 2023
Inventors: Pranshu Tiwari, Swarnalata Patel, Harish Bharti, Saurabh Trehan
Publication number: 20220326917
Abstract: A method, system, and computer program product for implementing automated software application generation is provided. The method includes scanning source code for identifying missing elements of hardware and software parameters associated with functional operation of software for development. The hardware and software parameters are analyzed and converted into configurable digital bins. An automation parameter is generated. The automation parameter is associated with portions of the software configured for automatic development and generation. Corrective actions associated with automating development of the software are generated based on the automation parameter and it is determined that the automation parameter is within a specified range of the portions. The corrective actions are executed with respect to development of the software and the software is generated.
Type: Application
Filed: April 13, 2021
Publication date: October 13, 2022
Inventors: Manohar Chintala, Swarnalata Patel, Rama Prasad Reddy Munagala, Pranshu Tiwari, Harish Bharti
Publication number: 20230401462
Abstract: A method for a predictive insulation condition-based recommendation policy includes determining a present insulation level for an insulation layer on a cable providing power to an electronic device based on a current leakage test. The method also includes determining a current leakage and a confidence score for the insulation layer on the cable providing power to the electronic device utilizing a supervised machine learning model. The method also includes generating a decision framework for the electronic device based on the current leakage and the confidence score in response to determining to perform an action based on the decision framework for the electronic device, the method also includes performing the action based on the decision framework to address the current leakage for the insulation layer on the cable providing power to an electronic device.
Type: Application
Filed: May 19, 2022
Publication date: December 14, 2023
Inventors: Saurabh Trehan, Pranshu Tiwari, Harish Bharti
Patent number: 10935970
Abstract: A method collects from at least one power measuring device of an environment power consumption data indicating active power consumption during a timeframe by an electrical device in the environment. The method also collects operating parameter data indicating at least one operating parameter under which the electrical device operates during at least a portion of the timeframe. The method performs, based on observing an increase in power consumption of the electrical device during the timeframe, assessing extents of contribution by potential contributing factors to the increase in power consumption, the potential contributing factors including time-based degradation of the electrical device and the at least one operating parameter. The method outputs, based on the assessing, an indication of an extent of contribution of degradation of the electrical device to the increase in power consumption.
 
Type: Grant
Filed: October 10, 2019
Date of Patent: March 2, 2021
Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
Inventors: Harish Bharti, Abhay K. Patra, Pranshu Tiwari, Saurabh Trehan
Patent number: 10558206
Abstract: A method collects from at least one power measuring device of an environment power consumption data indicating active power consumption during a timeframe by an electrical device in the environment. The method also collects operating parameter data indicating at least one operating parameter under which the electrical device operates during at least a portion of the timeframe. The method performs, based on observing an increase in power consumption of the electrical device during the timeframe, assessing extents of contribution by potential contributing factors to the increase in power consumption, the potential contributing factors including time-based degradation of the electrical device and the at least one operating parameter. The method outputs, based on the assessing, an indication of an extent of contribution of degradation of the electrical device to the increase in power consumption.
 
Type: Grant
Filed: May 9, 2017
Date of Patent: February 11, 2020
Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
Inventors: Harish Bharti, Abhay K. Patra, Pranshu Tiwari, Saurabh Trehan
Publication number: 20160224965
Abstract: An approach is provided for determining an optimal payment instrument. A computer in a system configured in one of four predetermined cloud-based models updates a database with details about account balances and rewards provided by a customer’s multiple accounts specifying payment instruments. A request is received from a near field communication enabled mobile device to purchase an item from a retailer. The request includes identifications of the retailer and the item. Based on the identifications, details about the account balances and rewards are retrieved from the database. Based on the details about the account balances and rewards, rules are applied and the optimal payment instrument is determined. A display of the optimal payment instrument to the customer is initiated. The optimal payment instrument is selected and a payment for the item by the optimal payment instrument is initiated.
 
Type: Application
Filed: February 4, 2015
Publication date: August 4, 2016
Inventors: Deepak Chauhan, Pranshu Tiwari, Saurabh Trehan
Patent number: 11645049
Abstract: A method, system, and computer program product for implementing automated software application generation is provided. The method includes scanning source code for identifying missing elements of hardware and software parameters associated with functional operation of software for development. The hardware and software parameters are analyzed and converted into configurable digital bins. An automation parameter is generated. The automation parameter is associated with portions of the software configured for automatic development and generation. Corrective actions associated with automating development of the software are generated based on the automation parameter and it is determined that the automation parameter is within a specified range of the portions. The corrective actions are executed with respect to development of the software and the software is generated.
Type: Grant
Filed: April 13, 2021
Date of Patent: May 9, 2023
Assignee: International Business Machines Corporation
Inventors: Manohar Chintala, Swarnalata Patel, Rama Prasad Reddy Munagala, Pranshu Tiwari, Harish Bharti
Publication number: 20200041991
Abstract: A method collects from at least one power measuring device of an environment power consumption data indicating active power consumption during a timeframe by an electrical device in the environment. The method also collects operating parameter data indicating at least one operating parameter under which the electrical device operates during at least a portion of the timeframe. The method performs, based on observing an increase in power consumption of the electrical device during the timeframe, assessing extents of contribution by potential contributing factors to the increase in power consumption, the potential contributing factors including time-based degradation of the electrical device and the at least one operating parameter. The method outputs, based on the assessing, an indication of an extent of contribution of degradation of the electrical device to the increase in power consumption.
Type: Application
Filed: October 10, 2019
Publication date: February 6, 2020
Inventors: Harish BHARTI, Abhay K. PATRA, Pranshu TIWARI, Saurabh TREHAN

Patent number: 11093229

Abstract: A failure rate model modeling a failure rate of a training functionality deployment in a training set of functionality deployments is constructed. The failure rate model is configured to receive functionality deployment data and output a corresponding failure rate prediction. Using the failure rate model, a set of functionality deployment failure rates is predicted, a functionality deployment failure rate in the set of functionality deployment failure rates corresponding to an upcoming functionality deployment. Using the set of functionality deployment failure rates, a deployment sequence of the set of upcoming functionality deployments is constructed to minimize a predicted overall failure rate of the set of upcoming functionality deployments. The deployment of each functionality deployment in the set of upcoming functionality deployments is caused, the deployment comprising activating the upcoming functionality program code according to the deployment sequence.

Type: Grant

Filed: January 22, 2020

Date of Patent: August 17, 2021

Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventors: Pranshu Tiwari, Harish Bharti, Naveen Narayanaswamy, Ram Prasad Reddy Munagala

Publication number: 20180356776
Abstract: A method, system and computer program product are disclosed for rewarding customers for sustained energy conservation. In embodiments of the invention, the method comprises establishing a baseline of energy consumption for a customer; defining a rewards program; and determining a reward for the customer, based on the rewards program, for maintaining the customer’s energy utilization in a given range compared to the baseline. In embodiments of the invention, the system comprises one or more processing units configured to operate as a file server for receiving specified energy consumption data from customers, and a business analytical engine. The business analytical engine stores a defined rewards program; establishes a baseline of energy consumption for one of the customers; and determines a reward for the one customer, based on the rewards program, for maintaining the one customer’s energy utilization in a given range compared to the baseline.
 
Type: Application
Filed: June 7, 2017
Publication date: December 13, 2018
Inventors: Harish Bharti, Abhay K. Patra, Pranshu Tiwari, Saurabh Trehan
Publication number: 20230245011
Abstract: Disclosed are techniques for training and utilizing a machine learning model for automatically accelerating technical incident triage. A historical dataset is combined from a plurality of sources corresponding to historical technical incidences and how they were resolved, including what persons or responders were involved in responding to the incident. This historical dataset is processed for input into a machine learning model which is trained to output a ranking of historical incidences based on similarity to an input incident. Machine logic then automatically selects a responder for the incident based on which available responders previously resolved the ranked historical incidences, selecting responders from the most similar historical incidences if they are available, and communicating information from the ranked historical incidences to assist in diagnosing and resolving the input incident.
Type: Application
Filed: January 18, 2022
Publication date: August 3, 2023
Inventors: Pranshu Tiwari, Swarnalata Patel, Harish Bharti, Manohar Chintala
Publication number: 20230245011
Abstract: Disclosed are techniques for training and utilizing a machine learning model for automatically accelerating technical incident triage. A historical dataset is combined from a plurality of sources corresponding to historical technical incidences and how they were resolved, including what persons or responders were involved in responding to the incident. This historical dataset is processed for input into a machine learning model which is trained to output a ranking of historical incidences based on similarity to an input incident. Machine logic then automatically selects a responder for the incident based on which available responders previously resolved the ranked historical incidences, selecting responders from the most similar historical incidences if they are available, and communicating information from the ranked historical incidences to assist in diagnosing and resolving the input incident.
Type: Application
Filed: January 18, 2022
Publication date: August 3, 2023
Inventors: Pranshu Tiwari, Swarnalata Patel, Harish Bharti, Manohar Chintala
Patent number: 9766693
Abstract: Provision of power scheduling framework for a plurality of virtual machines to facilitate energy savings in handing service instances includes configuring a first set of virtual machines in an active power mode and a second set of virtual machines in one or more other power modes, and managing, across a time period and based on service instance arrival, power mode configuration for the first set of virtual machines and the second set of virtual machines, the managing including initiating one or more transitions of at least one virtual machine between power modes of the plurality of power modes, wherein a transition energy cost is associated with transitioning from a lower power mode of the plurality of power modes to a higher power mode of the plurality of power modes, and wherein the managing accounts for transition energy cost in determining the one or more transitions.
 
Type: Grant
Filed: January 7, 2016
Date of Patent: September 19, 2017
Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
Inventors: Rajesh K. Saxena, Pranshu Tiwari, Vikram Yadav
Publication number: 20170199564
Abstract: Provision of power scheduling framework for a plurality of virtual machines to facilitate energy savings in handing service instances includes configuring a first set of virtual machines in an active power mode and a second set of virtual machines in one or more other power modes, and managing, across a time period and based on service instance arrival, power mode configuration for the first set of virtual machines and the second set of virtual machines, the managing including initiating one or more transitions of at least one virtual machine between power modes of the plurality of power modes, wherein a transition energy cost is associated with transitioning from a lower power mode of the plurality of power modes to a higher power mode of the plurality of power modes, and wherein the managing accounts for transition energy cost in determining the one or more transitions.
Type: Application
Filed: January 7, 2016
Publication date: July 13, 2017
Inventors: Rajesh K. SAXENA, Pranshu TIWARI, Vikram YADAV
Publication number: 20210224053
Abstract: A failure rate model modeling a failure rate of a training functionality deployment in a training set of functionality deployments is constructed. The failure rate model is configured to receive functionality deployment data and output a corresponding failure rate prediction. Using the failure rate model, a set of functionality deployment failure rates is predicted, a functionality deployment failure rate in the set of functionality deployment failure rates corresponding to an upcoming functionality deployment. Using the set of functionality deployment failure rates, a deployment sequence of the set of upcoming functionality deployments is constructed to minimize a predicted overall failure rate of the set of upcoming functionality deployments. The deployment of each functionality deployment in the set of upcoming functionality deployments is caused, the deployment comprising activating the upcoming functionality program code according to the deployment sequence.
Type: Application
Filed: January 22, 2020
Publication date: July 22, 2021
Applicant: International Business Machines Corporation
Inventors: Pranshu Tiwari, Harish Bharti, Naveen Narayanaswamy, Ram Prasad Reddy Munagala
Publication number: 20220138871
Abstract: A method, a computer system, and a computer program product for restoration time predictions and optimized recovery solutions is provided. Embodiments of the present invention may include selecting a dataset based on time series data. Embodiments of the present invention may include building one or more models, wherein the one or more models include a classification and regression tree model, a cross validated decision tree model or a bootstrap aggregating model. Embodiments of the present invention may include selecting a model for predictions using decision trees to reduce a variance of the model. Embodiments of the present invention may include rebuilding the model based on additional data attributes. Embodiments of the present invention may include determining an outage prioritization for each outage node. Embodiments of the present invention may include transmitting results of the outage prioritization.
Type: Application
Filed: November 5, 2020
Publication date: May 5, 2022
Inventors: Pranshu Tiwari, Atul Kumar Thapliyal, Harish Bharti, ABHAY KUMAR PATRA
Publication number: 20240370287
Abstract: Computer implemented methods, systems, and computer program products include program code executing on a processor(s) that ingests data from one or more computing environments, where the data is related to applications. The processor(s) identifies, based on utilizing topic modeling and latent semantic analysis of the data, homogenous applications among the applications, which include analyzing subdata handled by each application and functionalities of each application; the homogenous applications comprise similarities in the data and in the functionalities. The processor(s) determines overlapping data among the homogenous applications based on the topic modeling, the latent semantic analysis, and term frequency-inverse document frequency of terms in the overlapping data. The processor(s) selects, from the overlapping data, training data. The processor(s) utilizes the training data to calculate weights for disposition metrics and the metrics to predict the resource dispositions for the applications.
Type: Application
Filed: May 4, 2023
Publication date: November 7, 2024
Inventors: Pranshu Tiwari, Swarnalata Patel, Saurabh Trehan, Harish Bharti, Abhaya Kumar Sahoo
Patent number: 11501237
Abstract: A method, system, and computer program product for predicting optimal service characteristics to execute predicted IT tickets. The method may include identifying deployment characteristics for an operational system based on an architecture of the operational system. The method may also include receiving implementation data in response to past incident tickets based on the deployment characteristics. The method may also include identifying relationships within the implementation data. The method may also include predicting a volume of new tickets based on the implementation data and the relationships between the implementation data. The method may also include predicting a resolution time for high severity tickets in the volume of new tickets. The method may also include determining preferred service characteristics based on the volume of the new tickets and the resolution time for the high severity tickets.
Type: Grant
Filed: August 4, 2020
Date of Patent: November 15, 2022
Assignee: International Business Machines Corporation
Inventors: Pranshu Tiwari, Harish Bharti, Pradeep Dash, Santosh Patil, Swarnalata Patel, Saurabh Trehan
Publication number: 20220044179
Abstract: A method, system, and computer program product for predicting optimal service characteristics to execute predicted IT tickets. The method may include identifying deployment characteristics for an operational system based on an architecture of the operational system. The method may also include receiving implementation data in response to past incident tickets based on the deployment characteristics. The method may also include identifying relationships within the implementation data. The method may also include predicting a volume of new tickets based on the implementation data and the relationships between the implementation data. The method may also include predicting a resolution time for high severity tickets in the volume of new tickets. The method may also include determining preferred service characteristics based on the volume of the new tickets and the resolution time for the high severity tickets.
Type: Application
Filed: August 4, 2020
Publication date: February 10, 2022
Inventors: Pranshu Tiwari, HARISH BHARTI, PRADEEP DASH, SANTOSH PATIL, Swarnalata Patel, Saurabh Trehan
Patent number: 11810209
Abstract: A method, a computer system, and a computer program product for restoration time predictions and optimized recovery solutions is provided. Embodiments of the present invention may include selecting a dataset based on time series data. Embodiments of the present invention may include building one or more models, wherein the one or more models include a classification and regression tree model, a cross validated decision tree model or a bootstrap aggregating model. Embodiments of the present invention may include selecting a model for predictions using decision trees to reduce a variance of the model. Embodiments of the present invention may include rebuilding the model based on additional data attributes. Embodiments of the present invention may include determining an outage prioritization for each outage node. Embodiments of the present invention may include transmitting results of the outage prioritization.
Type: Grant
Filed: November 5, 2020
Date of Patent: November 7, 2023
Assignee: International Business Machines Corporation
Inventors: Pranshu Tiwari, Atul Kumar Thapliyal, Harish Bharti, Abhay Kumar Patra