Data Science - Manufacturing

Role & Responsibility:

- As an upcoming technology area, the AI/ML Digital lead will be entrusted with creating the foundation for an AI/ML based framework for analysing & insighting data from Manufacturing and SC domains for a Smart Data driven factory.

- In addition to creating AI/ML based solutions for data driven decision-making, this role will support establishing capabilities for an AI/ML based discipline, supported by data engineers, and a competency -based approach which can be leveraged across the enterprise.

A few examples of proposed initiatives and milestones are as below :

- Migrating from Condition based monitoring to Predictive and Cognitive monitoring.

- Planning the insights evolution from categorized quality to Predictive and Cognitive quality.

- Intelligent and real time Production Scheduling, to maximise energy and plant efficiency while minimizing set up gaps. Cognitive allocation of Manpower and resources, to maximise resource efficiency.

- Intelligent synchronization of manufacturing to support Demand forecasting and inventory optimisation.

- Smart Co-relation of 4 M parameters for root cause analysis, improving Quality, Energy and Inventory parameters etc.

- The proposed role will lead these projects and initiatives from the point of creating functional data driven templates for intelligent decision making in real time.

- Presently working on all above projects, and has data engineering and analysis projects in these areas. It is now proposed to leverage the AI/ML approach to all these areas, and generate commensurate benefits from AI/ML based insights.

- These and more such initiatives will need to be supported by an AI/ML framework for the creation of Prediction Models, Virtual twins, and Probability indices .Data models with Co-relations in a multi variate environment, as appropriate to the specific context (from the engineering view point), creating comparisons with physics based models and empirical models for best fit.

Key technical abilities:

- Expected to have excellent knowledge of computer engineering fundamentals, algorithms and data structures.

- Proficiency in using SQL, Mongo DB,

- Programming expertise in Python, or R/Julia, with hands on in C++ or Java with experience of using Matplotlib, Sci-Py Pandas, etc.

- Experience with Supervised and Unsupervised learning, and should have worked with specific instances of applying techniques such as Linear, Logistic Regression, MLNN, Decision Trees, K-NN, Naive Bayes, SVM Decision Forests, XGBoost, LightGBM. Should have demonstrated experience in building visualisations in Matplotlib, seaborn. Experience in deep learning is an advantage.

Machine learning & AI exposure :

- Proficient in Tensor Flow, Pytorch, scikit learn, and Google ML. Handling time series data, Supervised learning, Reinforcement learning.

Statistical Analysis:

- Good understanding of Mean, Standard Deviations, Distributions, Gaussian Distributions and regression analysis.

Data:

- Undertake to preprocess of structured and unstructured data. Analyze large amounts of data to discover trends, patterns and create hypotheses, related to manufacturing and supply chain conditions.

Model building:

- Develop and Scale up predictive models and ML Algorithms for Machine prediction, Process parameter wise prediction. Combine models through ensemble modelling. Model tuning, model comparisons, and model monitoring.