Intelligent manufacturing, leveraging advanced technologies for end-to-end digitization and automation could be the answer. Let us delve deeper.

Siloed Data Ecosystems, Lack of Real-time Visibility and Reactive Approach Hinder Efficiency

A large number of manufacturing units run on aging equipment causing 45% unscheduled downtime. Operations are disrupted in factories as asset health is not regularly monitored. Maintenance of equipment becomes a major challenge and 80% of the time production managers and factory staff are fixing issues after the damage is done. The ability to proactively apprehend and resolve incidents is limited, leading one in three manufacturers to spend more than 10% of their operating budget on maintenance alone. At the enterprise level visibility on plant health, operating processes, and floor level performance is hindered with siloed data and access to actionable reports, on time. This creates an environment of vulnerability, posing challenges in meeting safety and regulatory compliance requirements. The end result is sub-optimal productivity and sub-par performance impacting the bottom line.

Intelligence Powered Digital Next Manufacturing Accelerates Growth and Amplify Outcomes

The manufacturing sector has the opportunity to reinvent their production and supply chain, augment the workforce, and drive deeper engagement with their customers with advanced technology adoption. The ability to harmonize, analyze, and drive real-time insights from multiple data streams define the strength of a digital enterprise. No wonder, more than 50% of manufacturers are keen to invest in building Big Data capabilities. The need for real-time inline quality control and visibility into the end-to-end process is driving 80% of technology leaders in the sector to increase investments in digitization, Artificial Intelligence, and Advanced Analytics.

Cognitive supply chain and operations with AI-based advanced planning and automation improves the ability to pre-empt issues, sense the operational challenges to respond with agility and speed. This removes vulnerability from systems and improves the health of assets by over 40%, saving huge maintenance costs and reducing unplanned downtime by almost 50%.

Companies improve the bottom line with a 5-10% reduction in operating cost and increased earnings by up to 10%. The advantages are obvious, the question is how do manufacturers accelerate their journey to digital next future?

Roadmap to Digital Next Manufacturing��

As manufacturers embrace emerging technologies the focus must remain on creating a definitive roadmap to achieve desired goals. Here is an incremental approach, that equips technology leaders to chart a sustained path to digital enablement:

Roadmap for Digital Next Intelligent Manufacturing

Connect Plant and Enterprise – A��strong data foundation built on IT-OT convergence and integration across multiple data streams forms the bedrock of a connected organization. Implementing gateway solutions that provide multi-device support and build a machine to machine (M2M) connectivity enable real-time visibility, optimize capacity, and improve the quality of the output.

Conduct Prescriptive and Predictive Analytics –��Once the organizations gain the ability to harness data, they are enabled to leverage AI, ML, and advanced analytics to build predictive models and interactive dashboards. Real-time data management and AI/ML algorithms provide prescriptive analytics to improve overall equipment efficiency and reduce inventory. As they decrease the lead time for procurement there are substantial cost savings, equipping business leaders to optimize their budget.

Drive Cognitive Intelligence��– As manufacturers operationalize self-tuning AI/ML algorithms at the enterprise level, they gain the ability to contextualize intelligence. Processes are automated with a digital feedback loop and systems are equipped to take decisions with scripted algorithms. These advanced cognitive capabilities pave way for a responsive and flexible value chain and proactive risk management.

Augment Intelligence on Edge��–��Building on cognitive technologies, CIOs can adopt edge computing and advanced AI/ML modeling to connect products and services. The strength of their ML algorithms will create an ecosystem for autonomous decision making to keep the assets performing at peak. Augmented and Virtual Reality (AR/VR) adoption can bring them closer to floor level operations and create opportunities for product innovation. The technology prowess ultimately translates to enhanced productivity and efficiency, leading to improved margins.

Manufacturing Platforms of Intelligence

Under our , we offer Factory, Supply Chain, and Product Intelligence to equip manufacturers to build digital next manufacturing organizations that are cost-efficient, high performing, and more profitable. We can work with you to explore the use cases and assess the readiness with us through a discovery workshop and demonstrate value with a pilot MVP for a use case in 3 months.

91���� has been supporting manufacturers to gain streamline operations and gain a competitive advantage with Platforms of Intelligence. Our pedigree in CPG and manufacturing comes from our expertise and experience with our parent group ITC. We bring the advantages of Cloud, Big Data, Advanced Analytics, AI/ML, and Augmented Reality on a single platform, we accelerate the journey fully connected ecosystem. We leverage our various strategic partners like Microsoft, PTC ThingWorx, Anaplan, and Salesforce.

To know more, write to us.

Authors:

Amit Tulshiram Derkar
Senior Manager, DX Offerings & GTM

References

The post Digital Next Manufacturing Powered by Platforms of Intelligence appeared first on 91����.

Adopt ML-based optimal booking strategy to maximize the yield and revenues

Overbooking flights is a common practice in the aviation industry based on the expectation that some fraction of the booked passengers will either cancel their booking at the eleventh hour or fail to show up at the departure gate (commonly known as no-show). Accurate forecasts of the number of no-show passengers for each flight can help the airline to accrue more revenue by simultaneously reducing the number of spoiled seats and lowering the cases of involuntary denied boarding which produce significant cost penalty. Optimal booking policies seek to maximize the yield as a tradeoff between the revenue due to additional sales offset by the cost of any denied boarding that might happen. Flights flying fully occupied are not only of economic interest but are of ecological importance as well.

Implementation of relevant statistical/machine learning based algorithms may ensure an advanced and robust prediction of the no-show numbers for each scheduled flight, which can aid the airlines company to accept optimum number of additional bookings over the inventory capacity. The solution comprised of two steps – prediction of cabin-level no show rates using specific information on the individual passengers booked on each flight and later employing these predicted no-show rates along with features extracted from historically similar flights to arrive at the final no-show estimate.

Build a robust prediction framework to identify no-shows

At the beginning, individual passenger details from historical flights were utilized to capture their booking trend and understand their travel pattern. Different classification models were employed to capture the different bias-variance structures and then ensembled together to come up with a more robust prediction framework. Identification of individual learners that will eventually lead to lower misclassification is the key. Below is the mis classification table.

Here B2 are the cases where a passenger predicted as flown have actually flown. A1 are the cases where one predicted as no-show didn’t turn up. The passengers who have been predicted to be flown passenger but in reality didn’t turn up are A2. B1 are the passengers who have been predicted to not turn up but eventually did. Though both A2 and B1 cases are misclassified, an increased number of B1 cases will result into higher instances of denied boarding which will escalate the cost in the form of penalty.

At the next step some generalized linear model concepts were employed to come up with the estimated no show numbers. A GLM model for the unique combination of i^th Flight Market Departure Date is of the form:

g(y_i) = f(x_i) + ε_i

where g() is the appropriate link function applied on the response variable, that would help to model it as a linear function of the explanatory variables. In this case, f() is a linear function and ε is the corresponding error vector.

While experts will agree that individual passenger information for a particular flight is of immense importance, airlines usually accept bookings till few hours before departure, and hence any model based on the same data will be too volatile, and always be based out of incomplete data. Rather, we assume that the pattern of travelers in a particular Flight- Market would be similar in certain Days of Week. So, we utilize the cabin level no show rate from the first model as an auxiliary variable for the passenger level information in a GLM (Generalized Linear Model).�� We used GLM over other because of following reasons

• The response variable is a count variable (more than 0 but less than a certain number, which may vary from Market to Market)
• The variance structure of the error will guide on the selection of the appropriate link function
• The response variable might also be taken to be of type ratio (rate of no show) and may guide to selection of a totally different link function

A GLM yielding lower RMSE (Root Mean Square Error) values is considered as an appropriate prediction equation.

The General workflow can be summarized by the below diagram:

Improved inventory visibility and intelligent network planning for airlines

Some benefits reaped by the Airlines those have implemented Optimal Overbooking Strategy are:

• Reduction of spoiled seats as well as the number of involuntarily denied boarding that not only maximized the revenue of the airlines due to additional bookings but also curtailed additional cost due to compensation due to DNBs and gave a boost to the reputation of the company
• Accurate prediction teamed with historical booking pattern helped the client in network planning and inventory management

How scenario changes post pandemic

COVID-19 has changed how industries operate and Aviation is no exception. Most of the airlines worldwide had been grounded and very few are operating across limited routes. Some Indian airlines have started their domestic and international operations, but the number of flights deployed is just a fraction of their pre-pandemic frequency. Center for Aviation (CAPA) India has indeed stated that, “Indian Aviation Sector may lose $4 billion in FY 21.”* Currently, with limited transportation options, airlines is the only viable mode of transportation as people stranded in places away from home has an urge to return to near and dear ones.

In such trying situations having an Optimal Booking Policy at place can truly help sustain airlines. Though prediction of overbooking becomes more difficult in such an unprecedented situation, the airlines deems it to be even more important to have a robust and accurate demand forecast along with an Overbooking Strategy. Smart deployment of inventory along with the decision to set the Optimum Booking Limit, might help in the yield management more than ever.

New consideration of factors like instances of relaxation or implementation of sudden lockdown due to increasing cases of Corona outbreak at respective cities are important as this may result into cancellation of tickets for some passengers due to lack of transportation facilities giving room for overbooking.

Though one might say that the future of aviation looks bleak in current scenario, with the right changes applied at the right time to the existing model, we have the potential to maximize the revenue of an airline today and in the future.

We are building capabilities like these using AI/ML for optimal overbooking strategy in our platform of intelligence for airlines. Our platform can help airlines accelerate the value realization – 2% incremental PLF, 4% savings in fuel 3% upside in crew utilization and 2% improvement in TAT – using airline intelligence platform.

Stay tuned to know more about for .

Authors:

Devangana Dasgupta
Data Scientist, DATA

Akshay Vijay Medhane
Junior Data Scientist, DATA

Anindya Neogi
GM, Chief Data Scientist, Digital Experience

References

• “Passenger Based Predictive Modelling of Airline No Show Rates” – Richard D. Lawrence, Se June Hong, Jacques Cherrier
• “Forecasting and Forecast Combination in Airline Revenue Management Applications” – Christian Lemke, Bogdan Gabrys
• “Cancellation Predictor for Revenue Management applied in the Hospitality Industry” – R. van Leeuwen
• Categorical Data Analysis, Second Edition – Andre Agresti

The post Optimal Overbooking Strategy of Airlines using Statistical/Machine Learning Algorithms & their use in post pandemic situation – an expert view appeared first on 91����.

The post Product quality improvement in manufacturing using Machine Learning and Stochastic Optimization appeared first on 91����.

Platform-enabled transformations

Customer life cycle management is increasingly moving towards the cloud which integrates various functions and provides a 360-degree view of customer financials, including channel transactions, account opening and closing, default, fraud and customer attrition. Contact & Lead Management and Loan/Deposit Workflows are going completely paperless and contactless without the need for customers to visit branches. Advanced Analytics is providing customers with instant information on products along with pricing recommendations/optimization. It also provides the bank the vital ability to cross-sell/up-sell, create customer churn predictions and improve performance.

New trends like incentivizing retail customers for deposit/loan referrals are picking up. Various personalized offers and promotions are being made to customers through the use of Loyalty Points. Mobile Apps are being developed for Loyalty Management and Loyalty Points Redemption. Offer Management, Campaign Planning and Audience Management are integrated with Cloud-Based Architectures.

Social media and sentiment analysis provide actionable insights and quantifiable predictions about customer behavior. It also assesses brand awareness and can measure the success/failure of campaigns. Unstructured data analytics incorporates information from online discussion forums, social networks and call scripts to determine customer sentiment or market opportunities.

Insights about customer behaviors like shopping patterns, lifestyle, eating habits, credit history, spending habits etc. can be uncovered through multivariate descriptive analytics, as well as through predictive analytics. It can help improve a bank’s ability to segment, target, acquire and retain customers. Additionally, improvements in risk management and understanding the customers enable digital next banks to maintain and grow a more profitable customer base.

Risk analytics for credit risks, market risks and operational risks such as digital credit assessment, advanced early-warning systems, next-generation stress testing, credit-collection analytics etc., are performed by applying risk indicators to large datasets to detect risks that would otherwise remain hidden.

• Natural-language analytics are used to understand questions, context and semantics and analyze terabytes of data to identify and rank likely answers
• Algorithms are created to measure their own accuracy and feed that information back into the model to create self-improving predictive analysis
• Real-time analysis of data sources, such as financial markets, stock exchanges, or news is performed to gain risk insights

Leading use cases for banks

Kick-starting an effective use case

Implementing platforms of intelligence across the bank can be a daunting and potentially expensive prospect due to:

• Complex, heterogeneous technology architectures
• Operationally optimized but siloed processes and systems
• Data fragmented across multiple databases
• Constrained investment budgets with competing agendas
• Lack of skilled resources
• Perception that the available data lacks the quality to support analysis

All of these are genuine obstacles. However, it should not be assumed that analytical insight cannot be extracted until all the obstacles have been overcome. That road leads either to major programs striving to create perfect data that can answer any question or to an acceptance that any such effort is futile. Organizations do not necessarily have to solve all these issues to initiate an intelligent platform project and put it on the path to success.

A more pragmatic approach starts with selecting a critical question or objective, identifying the necessary data and recognizing that the data is not perfect. This allows the business to derive answers and information correlations with a corresponding confidence level. This approach does not replace the strategic architecture investment required to reach accuracy, but it provides a framework for business owners to control the level of their expenditures in a way that is proportionate to the benefit unlocked.

AI/ML based platforms of intelligence is the way forward for digital next banks across the world which can transform how banks analyze data, breaking down siloed decision making and improving overall customer experience. It can enable banks to respond quickly to disruption, propel the business forward and outmaneuver competitors in the market.

Author:

Abhiram Shankara
Lead Consultant, DATA

The post Digital Next Banks with AI/ML powered Platforms of Intelligence appeared first on 91����.