Blog | Growth Champion 2023 by Economic Times and Statista

Minfy is amongst India's Fastest Growing Companies as per a study by The Economic Times and Statista and is listed amongst India's Growth Champions 2022.

  1. Home
  2. >
  3. 2023
  4. >
  5. March

Minfy declared a Growth Champion 2023 by Economic Times and Statista!

Minfy is amongst India's Fastest Growing Companies (32nd) as per a study run by The Economic Times and Statista and is listed amongst India's Growth Champions 2023.

Economic Times and Statista pick and rank a select group of businesses that have perfected the art of gaining ground year after year. Aside from growth, key considerations are that businesses have real economic substance, sustainability, and legitimacy.



Thank you The Economic Times and Statista.

#Indiagrowthchampion2023 #betterfocusbettergrowth #bionicforabetterworld #economictimes

Click here to see the detailed rankings.

Blog | AI-ML in Demand Forecasting

AI-ML use cases in Demand Forecasting

  1. Home
  2. >
  3. 2023
  4. >
  5. March

Demand forecasting is a common use case of AI-ML. It can be used to identify areas of improvement and best practices that can help businesses improve its forecasting accuracy. It helps in production planning and determine incremental impacts of new initiatives, plan resources in response to expected demand, and project future budgets.

The following image depicts a simple forecasting cycle. It is all about using historical data, train forecasting model on it, generate forecast to make informed decisions. It’s a cyclic process — when new data arrives, we feed it back into the forecasting model and retrain it so that the model remains up to date.

Demand forecasting cycle

A frequent question we get from customers, consultants and other business experts is “What would you consider a good level of forecast accuracy?”

However, it may not be the most relevant question to raise, though. First of all, forecasting is seldom the objective in any retail or supply chain planning scenario. It is always a means to an end. A forecast is only useful if it helps us accomplish other objectives, like increased shelf availability, accurate production planning, less wastage.

While forecasting is a crucial component of all planning activities, it is only one cog in the machine, which means that other factors like changes in consumer behaviour, economic turndowns, historical data quality can have a significant impact on the forecasting accuracy.

What degree of forecast accuracy can actually be obtained depends on a number of factors. This is one of the reasons why comparing forecast accuracy between businesses, or even between products within the same business, is so challenging.

When integrating demand forecasting systems, it’s essential to understand that they are vulnerable to unpredictable situations. So, the demand forecasting machine learning models should be re-trained according to current reality.

With large forecast horizon, the probability that future demand will be impacted by developments that we are not currently aware of grows noticeably. Weather-dependent demand is a straightforward illustration. If we need to make decisions on what quantities of summer clothes to buy or produce half a year even longer in advance, there is currently no way of knowing what the weather in the summer is going to be.

On the other hand, if we were in charge of overseeing the replenishment of ice cream in grocery shops, we could utilize short-term weather forecast to estimate how much ice cream to transport to each location.


Important features to consider while training ML model for forecasting:



Seasonality refers to the regular and predictable pattern of demand for goods or services that occurs at certain times of the year, such as holidays, weather changes, or cultural events. What’s more, understanding of seasonality for each of your products can give you a head start in demand planning. With a good grasp of how much consumer demand varies, you can select the right demand forecasting models, focus on making plan adjustments and spot true outliers that require further attention.

For example, a business that sells winter clothing may experience a significant increase in demand during the winter months and a decrease in demand during the summer months. Similarly, a business that sells ice cream may experience higher demand during the summer months and lower demand during the winter months.

Weather data feature effect

Demand will typically vary from region to region, depending on local calendars and weather. You may need to segment your data geographically to identify seasonality patterns. At the same time, aggregating forecasts to a less granular level — product category instead of product, for example — may make it easier to distinguish seasonal patterns from random noise.

Rather than including only historical time series and demand, incorporate weather data into a machine learning model for demand forecasting as the model can account for the impact of weather on consumer behaviour and give better predictions accordingly. This can lead to help businesses make better decisions about inventory management, marketing strategies, and other aspects of their operations.

Furthermore, weather data can be used to create more granular and accurate models by breaking down the weather into different variables such as temperature, precipitation, humidity, etc. This allows for a more nuanced analysis of how weather affects demand for different products or services.

The following code snippet reads weather data from a csv file and pre-processes it by resampling to daily frequency and imputes the missing values by linear interpolation imputation technique, so that it can used with the historical time-series data to train a forecasting ML model.

Code snippet to pre-process weather feature


For certain use cases, location is an important feature when training a demand forecasting ML model.

Competition & Vendors: The location of a business and its associated vendors location where a particular product is being sold can play a vital role. Also, if there are many similar businesses in the area, the demand for a particular product or service may be lower than in an area with few or no competitors.

Demographics: The demographics of an area may play a crucial role in determining the demand for a product or service. Different age groups, income levels, and cultural backgrounds have different preferences, and this can be reflected in demand patterns.

Infrastructure: The availability and quality of infrastructure such as roads, transportation, and logistics can also affect demand. Areas with poor infrastructure may have lower demand due to difficulties in accessing or transporting products.


Sales patterns can vary significantly during holiday seasons due to changes in consumer behaviour, such as increased shopping activities, changes in preferences, and higher demands for specific products or services.

Including holiday data into the machine learning model helps to improve the accuracy of sales forecasts during these periods, which is essential for businesses to plan their operations and inventory management. Additionally, taking holidays into account in the model can also reveal important insights into customer behaviour and preferences, which can help businesses tailor their marketing strategies to better engage with their target audience.


One of the benefits of including trends in demand forecasting algorithms is the ability to anticipate market changes and respond accordingly. For example, if a trend analysis indicates a shift in consumer preference for a particular product, the company can adjust their production and marketing strategies to meet the new demand. Additionally, trend features can help identify whether changes in demand are due to short-term fluctuations or long-term changes in the market.

Moreover, trends can also help identify potential bottlenecks in the supply chain, which can help companies optimize their inventory levels and reduce waste. By accurately forecasting future demand, companies can avoid stock outs or overstocking, which can lead to significant losses.

Let’s say you have a dataset that contains daily temperature readings for a specific location over a period of four years. You want to include a feature that represents the trend in temperature over time. One way to do this is to calculate a moving average of the temperature readings for each day, using a window size of, say, 30 days. This will give you a new feature that represents the average temperature over the past 30 days, which can be a good indicator of whether the temperature is trending up or down.

Code snippet to pre-process Temperature trend feature


Above code snippet, reads the ‘temperature_data.csv’ file which contains a column called ‘temperature’ that has the daily temperature readings. The rolling () function is used to calculate a rolling average of the temperature column with a window size of 30. The resulting rolling average is then added as a new column to the data frame, with the name ‘temp_trend’.

Machine learning models/tools that can be used for demand forecasting

• Amazon Forecast: Amazon Forecast is a time-series forecasting service based on machine learning (ML) and built for business metrics analysis. Amazon Forecast uses machine learning (ML) to generate forecasts with just a few clicks, without requiring any prior ML experience. If you have a time-series forecasting use case and you don’t have much experience in machine learning, Amazon Forecast is likely a good choice.

• Amazon SageMaker:It is a fully managed machine learning service. With SageMaker, data scientists and developers can quickly and easily train and test machine learning models, and then directly deploy them into a production-ready hosted environment. If you require more flexibility and control over your machine learning models, add and create features, built relationship between the features then Amazon SageMaker can be a better fit.

• Amazon Deep AR: The Amazon SageMaker Deep AR forecasting algorithm is a supervised learning algorithm for forecasting scalar (one-dimensional) time series using recurrent neural networks (RNN).

• ARIMA: An autoregressive integrated moving average model is a form of regression analysis that gauges the strength of one dependent variable relative to other changing variables. The model’s goal is to predict future data points by examining the differences between values in the series instead of through actual values.

• SARIMA: Seasonal Autoregressive Integrated Moving Average, SARIMA or Seasonal ARIMA, is an extension of ARIMA that explicitly supports univariate time series data with a seasonal component. It adds three new hyperparameters to specify the autoregression (AR), differencing (I) and moving average (MA) for the seasonal component of the series.

• LSTM’s: Long Short — Term Memory is a type of neural network architecture that is commonly used for time series forecasting. LSTM networks are particularly useful for dealing with time series data that have long-term dependencies or complex patterns. The basic idea behind LSTM is to use a series of memory cells to remember and store important information about the input sequence, and then use this information to make predictions about future values in the sequence.


Demand forecasting is a crucial aspect for any business operation, and the use of machine learning algorithms can significantly improve the accuracy and efficiency of the forecasting process. By considering historical data, current trends, and other relevant factors, businesses can better understand their customers’ needs and optimize their production and inventory levels to meet demand while minimizing the loss.

You should ensure that the process of re-training your ML model with new data happens at continuous intervals. The goal of model retraining is to adapt the model to changing patterns in the data, or to improve its accuracy as new data becomes available.

Retraining a model is important because machine learning models are only as good as the data they are trained on. It can also help to address issues such as model drift, where the performance of a model deteriorates over time due to changes in the data distribution.

In conclusion, demand forecasting using machine learning is not a one-time process, but a continuous effort that requires ongoing monitoring and retraining of models on latest available data. By leveraging the latest technologies and insights, companies can stay ahead of the curve and thrive in today’s dynamic and rapidly evolving marketplace.

 - Author Yasir Ul Hadi






Blog | AI in Energy & Resource Industry

Revolutionizing the Energy and Resources Industry: How AI is Reshaping the Future of Sustainability.

  1. Home
  2. >
  3. 2023
  4. >
  5. March

Revolutionizing the Energy and Resources Industry: How AI is Reshaping the Future of Sustainability.

The energy sector has never been sceptical in adopting new technologies in the past. Be it digitization of the power generating infrastructure, exploring renewable sources of energy or innovating the aged infrastructure of transmission and distribution, it has always embraced these challenges and adopted well.  However, challenges such as high carbon emissions, inefficient energy management, aging infrastructure, etc still needs to be addressed to build a more sustainable future. These challenges can be dealt with innovation and collaboration amongst stakeholders from different domains.

Artificial intelligence (AI) has the potential to transform the energy sector as it can address a wide range of problems that the sector is currently facing. The industry has already started to embrace the power of Artificial Intelligence (AI) to drive efficiency, reduce costs, and minimize carbon emissions.

According to Precedence Research, the global artificial intelligence (AI) in renewable energy market size is projected to surpass US$ 75.82 billion by 2030 and is expanding at a CAGR of 27.9% from 2021 to 2030.

In this article, we will explore some of the ways in which AI can help address the challenges that the energy sector is facing. Overcoming these challenges can create a more sustainable future for all. But before that, let's have a quick look at some of the key challenges in the energy sector.

1. Centralized power generation – majority of the global grid network relies on a few power generators making the entire infrastructure highly centralized. Though we have done a great job building such a great infrastructure, large, centralized networks have their own challenges. Some of these are high loss of energy from generation to distribution, high carbon emissions as most of these power plants are based out of non-renewable sources of energy, low power conversion rate, etc.

2. Challenges while transitioning to renewable sources – as the share of renewable sources of energy increases at a fast pace, integrating it with the existing grid system without destabilizing the grid, is quite challenging. Not only that, maintaining balance of demand and supply gets even more challenging as outputs of renewable sources of energy depends on external factors such as weather.

3. Energy Loss and theft: Improving energy efficiency is a key challenge for the energy sector. There is a need to reduce energy waste and increase the use of renewable energy sources to meet the growing demand. Theft and unethical usage of power is again a key concern causing losses to the power suppliers.

4. Ageing infrastructure: Much of the energy sector's infrastructure, such as power grids, equipment and pipelines, is ageing and in need of upgrades. This can be costly and can also create reliability issues.

Now, let’s see how AI can be leveraged to address these challenges in the energy sector.

Renewable energy forecasting: As the share of renewable energy is increasing at a fast pace, integrating it with the existing network is challenging. One of the key reasons is the fact that the production of energy from these sources is not consistent (solar and wind power plants, their power output depends upon external weather factors). Because of this, often the forecasts are inaccurate leading to either blackouts or wastage of surplus energy. Hence, accurately forecasting the output of wind, solar, and other renewable energy sources is essential for managing energy supply and demand. AI algorithms can be used to analyse historical data and real-time weather patterns to predict renewable energy output with greater accuracy, allowing grid operators to balance supply and demand more effectively.

Demand forecasting: AI algorithms can be used to forecast energy demand in real-time, helping grid operators to manage supply more efficiently and prevent blackouts. This can help balance supply and demand, reduce the need for backup power, and improve the overall reliability of the energy grid. This also helps in accurately planning purchase of power for different durations like intraday, short term, and long term. Other than demand forecasting, price forecasting is another important use case that helps in creating more informed power purchase agreements.

Let's have a look at the high-level architecture for demand forecasting using AWS services. With the help of AWS services like Forecast, Quicksight, Sagemaker, etc we can quickly build and deploy AI solutions at scale. Also, to make things easy, let’s assume that we are trying to forecast power demand for a utility.

The architecture consists of four broad functional components —

Data Preparation – First step is to prepare the demand data or the target time series, along with this we may add related time series data such as weather data, income data, and other metadata. Once this data is prepared for training, the data is uploaded to a S3 bucket from where it is used for machine learning and data visualisation tasks.

Model Training – Once the data pre-processing is done, use AWS Forecast to create a predictor or train a custom ML model using a Jupyter notebook with SageMaker. Once the model is trained, endpoint configuration is set and the endpoint is deployed to generate forecast. Amazon CloudWatch is used to monitor the model training and its metrics and Amazon SNS (Simple Notification Service) is used to send or receive notifications for training related events.

Forecasting/Inferencing – we can write a simple Lambda function that is triggered by S3 event notification. Whenever new data is uploaded to that S3 bucket, the Lambda function will be triggered to generate a forecast through AWS Step Function workflow which is a service to orchestrate tasks. All we need to do is upload data to the S3 bucket.

Data Visualization – Once a forecast is generated, the results and related metrics are then saved to the s3 bucket and Amazon Athena is used to query this data. After querying the results, Amazon Quicksight is used to visualize forecast results.

Predictive maintenance: Failure of electrical equipment leads to all sorts of undesirable conditions. These failures can lead to electrical safety hazards, unplanned downtime, increased maintenance costs, decreased equipment reliability and life span. Old predictive maintenance techniques were reliable in past but are now outdated. With reduced costs of sensors in the industry, AI solutions such as predictive maintenance can replace these old solutions and can transform entire system’s efficiency.

Predictive maintenance uses machine learning algorithms to predict when equipment is likely to fail, based on historical data and real-time sensor readings. By identifying potential issues before they occur, maintenance teams can schedule repairs during periods of low demand, reducing downtime and costs. Also, equipment running under healthy and efficient conditions will have longer lifespan, and reduced running costs.

Energy optimization: AWS’ collaboration with Carbon Lighthouse to reduce carbon emissions by optimizing energy usage with the help of Artificial Intelligence is a good example of energy optimization. AI solutions can be deployed to identify unoptimized scenarios and make them more efficient. Energy optimization involves using AI algorithms to optimize energy consumption across various processes, such as heating, cooling, and lighting. These algorithms learn the underlaying pattern and works to reduce power consumption by the overall process. This leads to reduced energy waste, improved efficiency, and lower costs.

Fraud Detection: Electricity theft and fraud is not a new challenge, it has been there for a long time and fraudsters keep innovating newer methods of unaccounted power consumption. However, with the help of new technology such as AI, more robust anti-theft systems can be built to help minimize such events. AI can be used to detect fraudulent activity in the energy sector, such as meter tampering or theft of service. By analysing usage patterns and other data, AI can identify suspicious behaviour and alert utilities to potential problems, helping to reduce losses and improve overall revenue.

Power Management – Wastage of power due to unoptimized solutions or human ignorance is a key concern when it comes to power management. When AI is integrated with sensors (IoT), we get a complete power management toolkit that can be installed anywhere from residential houses to commercial buildings. These AI powered smart solutions can help companies to cut down their utility bills by a significant amount. Till we have “arc-reactors” powering our buildings like Tony Stark’s NYC lab, we need an AI to at least manage the power of our buildings.

The energy sector is facing significant challenges in the modern world, including high centralization, transitioning to renewables, energy losses and thefts, and ageing infrastructure. To overcome these challenges, the industry must embrace innovation and collaborate with stakeholders to create sustainable solutions. AI is one such innovation that can help reduce carbon emissions, optimize energy production and distribution, manage renewable energy sources, and improve energy efficiency. By leveraging the power of AI, the energy sector can create a more sustainable future for all, while also ensuring reliable and secure energy supplies. The road ahead may be challenging, but with innovative solutions and collaborative efforts, we can build a brighter and cleaner future for generations to come.

Author Rishi Khandelwal

Blog | AI in Chronic Disease

How AI is Changing the Game in Chronic Disease Care

  1. Home
  2. >
  3. 2023
  4. >
  5. March

How AI is Changing the Game in Chronic Disease Care

As the world continues to grapple with the ever-increasing burden of chronic diseases such as diabetes, heart disease, and cancer, it becomes increasingly apparent that traditional methods of disease management and prevention are no longer sufficient.

According to the World Health Organization (WHO), chronic diseases account for almost 40 million deaths globally and are projected to increase to 57 million by 2020[1]. Furthermore, 80% of all deaths from chronic diseases occur in low- and middle-income countries[2].

The need for innovative solutions that address both the medical and lifestyle factors contributing to chronic disease development is paramount. The complexity of this problem requires a multi-disciplinary approach that takes into account the social, economic, and environmental determinants of health, making it a challenging issue to tackle.

The Challenge

Managing chronic diseases such as diabetes, heart disease, and cancer poses a considerable challenge due to the requirement of ongoing medical treatment and management. This ongoing care can be a significant burden for patients and their families, as it often entails frequent visits to healthcare providers, regular medication consumption, and the adoption of lifestyle changes to manage the disease. Furthermore, lifestyle factors, such as tobacco use, unhealthy diet, and physical inactivity, not only contribute to the development of chronic diseases but also make them more difficult to manage. These changes can be challenging and may require support from healthcare providers, family, and community resources. Addressing the challenges associated with chronic disease management necessitates a holistic approach that addresses both medical treatment and lifestyle factors.

In addition to the burden on patients and families, the rising cost of chronic disease management in India is a major concern for healthcare systems and policymakers. The economic cost of chronic diseases in India, which was around $100 billion just five years ago, is expected to continue to rise in the coming years, further exacerbating these challenges.

The Solution

Preventing the onset of chronic diseases such as diabetes, heart disease, and cancer is a crucial approach to addressing the challenges associated with their treatment and management. However, the overburdened and understaffed nature of healthcare systems makes it difficult to provide the necessary care and support to prevent these diseases from developing in the first place. To more effectively address these challenges, advanced tools and technologies are required to aid and better equip healthcare providers. Artificial Intelligence (AI) is one such tool that has the potential to revolutionize the management of chronic diseases.

Artificial Intelligence technology-based solutions can be leveraged to improve the prediction, diagnosis, and treatment of chronic diseases by providing healthcare professionals with more efficient and accurate tools. Utilizing AI in chronic disease management can also help to reduce costs and optimize resource utilization. It can assist healthcare professionals in making more informed decisions, identifying at-risk patients, and providing more personalized care.

AI-driven solutions have the potential to bridge the gap between the rising costs and the increasing burden of chronic diseases, making them more accessible and affordable for the population. By incorporating AI-driven solutions, healthcare systems can improve patient outcomes and create a more sustainable healthcare system.

There are several Artificial Intelligence technology-based solutions that are currently being used or have the potential to be used in chronic disease management and prevention:

1. Predictive modeling

Predictive modeling in chronic disease care can be used to identify individuals at high risk of developing a chronic disease and predict outcomes through analyzing patient data such as electronic health records. It can be helpful in chronic disease care by identifying individuals at high risk of developing a chronic disease, predicting outcomes, and guiding the treatment and preventive care decisions.

• Identifying high-risk patients: Machine learning algorithms can be trained on large amounts of data, such as patient EHRs, to identify patterns and predict outcomes.
• Predicting progression: Predictive modeling can be used to predict the progression of chronic diseases such as diabetes by analyzing data such as blood glucose levels and medication use.

2. Early detection

Early detection in chronic disease care refers to identifying the presence of a disease or the risk of developing an infection at an early stage, before the onset of symptoms, or before the disease has progressed significantly. Early detection can be beneficial in several ways:

• Improving outcomes: Early detection can lead to more effective treatment and management of chronic diseases, which can improve outcomes for patients. For example, if a patient with diabetes is diagnosed early on, they can take steps to control their blood sugar levels and prevent the development of complications.
• Reducing costs: Early detection can reduce the costs associated with chronic diseases, as treatments and management strategies are more effective when they are initiated early on.
• Increasing access to care: Early detection can increase access to care for patients, as they are more likely to be diagnosed and treated before the disease progresses and becomes more difficult to manage.
• Reducing the burden on healthcare systems: Early detection can also help to reduce the burden on healthcare systems, as patients with chronic diseases that are diagnosed early on are less likely to require hospitalization or other intensive care.
• Improving the quality of life: Early detection can improve the quality of life for patients, as they can take steps to manage their disease and prevent complications before they occur.

3. Personalized treatment

Personalized treatment or I would prefer to call it precision medicine, is an approach to healthcare that takes into account individual differences in genes, environment, and lifestyle. In the context of chronic disease care, personalized treatment can be helpful in several ways:

• Tailoring treatment to individual needs: Personalized treatment can improve treatment outcomes by taking into account individual differences in genes, environment, and lifestyle.
• Reducing side effects: Personalized treatment can reduce side effects by avoiding treatments that are unlikely to be effective or that may be harmful to the patient.
• Optimizing patient outcomes and reducing costs: Personalized treatment can improve patient outcomes by providing the right treatment for the right patient at the right time, thus reducing the risk of disease progression or complications, and reducing healthcare costs by avoiding unnecessary treatments.

4 Remote monitoring

Patient monitoring is the ongoing measurement and tracking of a patient’s health status in order to detect changes and respond accordingly. Supplementing patient monitoring with Artificial Intelligence can enhance the benefits of patient monitoring in chronic disease care by providing real-time data analysis, personalized treatment plans, and early warning signs.

• Identifying changes in health status: AI algorithms can analyze patient data in real-time and generate alerts for abnormal readings or trends, which can help identify changes in a patient’s health status and allow for early interventions.
• Monitoring treatment effectiveness: AI can process large amounts of data and identify patterns that would be difficult for a human to discern, which can be used to adjust treatment plans as needed. This can help to determine the effectiveness of a treatment plan by tracking changes in a patient’s health status over time.
• Improving communication between healthcare providers, patient outcomes, and engagement: AI can provide healthcare providers with real-time data analysis and personalized treatment plans, which can improve communication between healthcare providers and make treatment decisions more informed. Additionally, AI can provide early warning signs and personalized feedback and advice to the patients to better understand their health status and take action to improve it.

The Conclusion

AI is rapidly changing the paradigm of chronic disease care by leveraging its ability to process large amounts of data, identify patterns, and make predictions. Its potential to reduce mortality rates, alleviate the economic burden on healthcare systems, and improve the overall quality of life for individuals affected by chronic diseases, has been widely acknowledged by healthcare providers, policymakers, and community organizations.

From predictive modeling and patient monitoring to clinical decision support, AI is providing new and innovative ways to manage and treat chronic diseases. By utilizing sophisticated algorithms such as machine learning, deep learning, and natural language processing, AI is able to extract valuable insights from complex data sets. This can aid in the development of actionable strategies for disease management and prevention by healthcare providers, policymakers, and community organizations.

As AI technology continues to evolve and become more sophisticated, it has the potential to revolutionize chronic disease care, leading to better outcomes and a higher quality of life for patients. With its ability to provide real-time data analysis, personalized treatment plans, and early warning signs, AI is poised to change the game in chronic disease care, making it more efficient, effective, and personalized.

— Author: Gaurav Lohkna


P.S.: For a more detailed and technical analysis of AI-based solutions in Chronic Kidney Disease management (one of 13 major chronic diseases), kindly await the publication of my upcoming study/article. It provides an in-depth examination of cutting-edge technologies, algorithms, and methodologies in chronic care.

[1] World Health Organization (WHO). Noncommunicable diseases: Mortality
[2] World Health Organization (WHO). Global health observatory data.

This website stores cookie on your computer. These cookies are used to collect information about how you interact with our website and allow us to remember you. We use this information in order to improve and customize your browsing experience and for analytics and metrics about our visitors both on this website and other media. To find out more about the cookies we use, see our Privacy Policy. If you decline, your information won’t be tracked when you visit this website. A single cookie will be used in your browser to remember your preference not to be tracked.