Not just manufacturing, other industries also (for instance road and bridge
maintenance) stand to gain a lot from predictive maintenance techniques. Instead of
correcting equipment failures and breakdowns once they occur, predictive maintenance
prevents these problems from ever happening by using intelligent algorithms to
predict equipment failure.
In the absence of a preventive strategy (Run to Failure Maintenance), what is even
more costly than expensive machine repair/replacements, is extended down times which
impact organizations’ reputation and bottom line significantly.
Traditional preventive maintenance strategies tend to be cautious, leading to
additional and sometimes unnecessary expenditure and downtime.
With predictive maintenance it is possible to minimize both high repair costs and
unnecessarily maintenance. Both past trends and real time data are used offer
customers the optimal maintenance choice. Data used would come from a host of
sources: manual inspection records, weather information, usage data, manufacturers’
specifications, other environmental data as well as programmable controllers and a
plethora of IOT devices
Predictive maintenance techniques would also decide the spare parts stocking
policies, line up the vendors for just-in-time delivery of expensive parts and could
tie up with the production schedules.
Quickly understand customer issues and reduce the time to service
The customer service departments of e-commerce businesses are often inundated by complaints and
service requests. They need to maintain teams who study the emails/ help desk records and
internally distribute them for quick resolution.
A NLP solution could process the documents, search for key words, understand the type of
complaints and efficiently build sets for different functional departments to study and resolve.
This would reduce the requirement for manual segregation of the requests and complaints as well
as faster resolution, thereby increasing customer satisfaction. Structured data sets would be
built for continuous AI/ statistical analysis leading to the following:
Identify problematic products and initiating remedial product composition and
quality control steps to resolve the problems
Identify the organizations’ vendors who regularly appear and initiate steps for
Identify customers who complain regularly requiring telephonic/ videoconferencing
meetings to resolve possible misunderstandings
Identify behavioral anomalies indicating a chance of fraud
Identify training needs of service personnel.
Spot defects during manufacturing
As we know image recognition is already helping manufacturers around the world to
improve their quality control processes. The cameras set up for this reason search
and spot defects predefined in the programs.
AI can revolutionize image recognition based defect detection by using machine
learning. Images are mapped to defects later identified to enable the algorithms to
find clusters and relationships which were not known initially.
The systems would learn to overlook image differences caused by extraneous factors
like dust. As a result the separate quality testing needs could be substantially
reduced and to be able to continually predict product quality over multiple stages
of processing. This would enable course correction at any stage and ensure that the
final finished products satisfy all quality criteria.
Since processes are automated they are accurate, consistent and scalable and not
dependent on manual effort. This improves organization efficiency and profitability.
Improve compliance in hot metal production in blast furnaces
Artificial Intelligence techniques have been effectively used to set standards and check
compliance in hot metal production.
The techniques are used to build relations between masses of historical machine sensor data and
the output quality information available from the factory registers and product quality reports
produced by quality audit/ factory technical departments. The ranges for individual parameters
in which the output quality are compliant are established, and measures are implemented to
ensure that the blast furnace process control equipment work within the established ranges. The
resultant benefits are manifold:
Consistent quality in output hot metal
Improved overall productivity
Consequent improved sales figures and bottom line
Utilization of AI to derive the desired parameter values to control the blast furnace and other
iron making operations help in achieving automation and increasing operating efficiency
This is just one of the many areas and many industries where AI techniques can be used to
improve efficiency in factory operations.
Customer segmentation through transactions and clickstream data
In today’s environment e-commerce organizations carry a wealth of information about
their customers. Apart from demographic data possibly picked up at the time of
registering customers, transaction data, browsing patterns, mails and social media
behavior are all there. In a world where the customer is king, the ability to
capture and use these customer insights is critical to shape products and solutions
which exceed customer expectations.
Utilizing clickstream and supplementary customer data, models for customer
segmentation and behavior modelling can be built which can help track customers’
movement among different segments over time, accurately predict their future
behaviors, determine their LTV and make recommendations for improving customer
marketing that human marketers are unlikely to spot on their own.
Information on how long the customers linger on the e-commerce site, their browsing
patterns, what they end up purchasing, whether and how many times they contact
customer service, and can tell us a lot about buying habits and preferences. Though
most organizations capture a lot of data, few put it to meaningful use. Using the
information, together with whatever demographic data is available, one can create
customer segments and then apply advanced analytics to understand the needs and
possible next actions of the key segments. One particular organization relies on
customer data from smartphones and fitness wearables, such as sleep, mobility, and
communication patterns, to help make clinical assessments and diagnosis. Once an
organization can effectively mine its information base, such innovative solutions
HR analytics solution using Chatbot and NLP
Large industrial companies have thousands of employees with differing levels of
understanding computer usage and little or no knowledge of the HR rules and policies
which get implemented from time to time. These companies often look to intelligently
automate support functions, so that the support teams are lean and are free to
concentrate on strategy rather than get bogged down with routine issues.
An analytics tool for HR which uses a chatbot to engage with employees can be built
to address regular HR issues. A NLP processor would parse employee chatbot queries
and emails, auto-respond / auto-forward as required. The processor would have access
to store with answers to all the FAQs that the HR department needs to address. This
store could be intelligently updated with queries and manual responses which the
tool could not initially auto-respond to. The NLP engine would classify employee
messages/complaints/requests by type, by sentiment and seriousness. All the
information generated and status of response to complaints and queries could be
stored in a HR Analytics dashboard.
The HR department is just one example where a Chatbot with a NLP processor could be
gainfully used. The same approach could be used by the purchase department to
converse with vendors and would be vendors, by the service department, to respond to
customers with service complaints, by the Shares department, to interface with
Deep Learning for Thermoplastic Composites
Composites are increasingly being used in the automotive and aerospace industries because they are light (compared to metals), have superior performance at considerably lower weight and can handle higher temperature ranges. Multi-Fiber systems (Carbon-Glass; Glass-Polypropylene; Carbon-Nylon; etc.) are used in this process. For the thermoplastic composite to be consistent quality, it is necessary for the mix of fiber and thermoplastic to be uniform across the cross section of the composite material.
Deep Learning techniques have successfully been used to process images of the cross sections of the fibers to determine both whether the reinforcement has been uniform across the cross section and the percentages of different types of fiber and thermoplastic. The results need to be mapped to the set standards to determine the quality of the composite material.
Image processing and deep learning techniques are being used in many applications across multiple functions and thermoplastic composites is just a niche example. Uses are found in security and surveillance, medical treatment, manufacturing, automotive and gaming, to name just a few.
Root cause analysis from customer feedback
Root Cause Analysis Is required in cases where a number of situations where failures, performance issues, service complaints and other complaint issues are recorded, often in textual form, at multiple points:
Call centre records
Social media platforms
It is necessary to identify the actual problem areas, determine causes and initiate remedial strategies.
Applying NLP techniques on text data would create structured data sets and values, which could then be analyzed to identify the real problem areas. Text analytics using NLP can detect behavioral signals (emotion, empathy, communication). The techniques can be used to mine the interactions and to understand customers’ ideas, expectations and unfulfilled needs. We would then be able to:
Identify the Root Causes.
Identify customers most likely to stop buying again
Improve sales by understanding customer requirements, and mapping them with products