Introduction
The prospect of clean electrical energy generation has recently driven companies to make massive investments on renewable power, which has subsequently affected the operations and profits of existing traditional thermal power plants. Moreover, the coal quality and supply chain in India has severely impacted the bottom line of thermal power plant operators. This has necessitated robust supply sourcing strategy and improving the efficiency of such plants beyond the paradigm of supercritical technology. Power generating companies are looking to optimize their entire coal supply chain, with the most critical aspects being sourcing, blending, and firing, to achieve the ultimate objective of reducing generation cost per unit of power generated.
Problem Definition
One of our esteemed clients, a leader in the power industry, has indigenously developed an excel-based model, which they call ‘Decision Support System’ to optimize fuel costs and achieve maximum utilization. This model aims to integrate the four most critical components of the supply chain, viz fuel sourcing, coal handling, operations and commercial on to a single transparent platform to make informed decisions. The model has three components, namely demand forecasting (user input), optimized coal sourcing and blending, and optimized coal firing. Inputs corresponding to various transactional and operational parameters / constraints are fed into this model manually for optimized results, which are highlighted on some dashboards for the concerned departments for their actions.
The client has benefited from this model immensely as they have been able to realize cost savings to the tune of 30% in the past using this model.
To make this solution more robust and efficient, the client wanted Mindtree NxT to convert the excel model as-is into a packaged Python solution, which will manage three concerns.
- Excel Solver-based logic has a limitation on the number of variables that can be processed at a time.
- Feed the input data and click one button to get the optimization results, instead of the present way of uploading info and clicking on multiple buttons to arrive at the results.
- Enable an option to integrate the Python-based solution with SAP data so that the solution can be automated in the future.
Beyond the points coined above, there is another benefit in the form of live dashboards showing real-time model predictions with regards to sourcing and firing. The operations team need this visualization to align their actions keeping in line with the overall objective of optimizing the cost per unit of generation.
Download white paper to know more
