Data Strategy for IoT and Industry 4.0

Overview

When people talk about industry and manufacturing, their heads often turn to large scale production with huge machines and a massive pool of people doing a multitude of jobs. Although many industrial companies and factory settings have been using forms of digital technology for some time now, only in recent years has this started to become more mainstream. There has been a realization that using artificial intelligence (AI) and internet of things (IoT) devices through the supplier chain, can have amazing effects on efficiency and cost savings.

This article looks at how data, AI and IoT can superpower industry, the type of infrastructure needed to make it work and the steps required for it to happen.

What is AI and IoT?

When you hear about AI in the media, it often sounds very futuristic, but in reality AI is not at the level that movies or television might lead us to believe. Whilst companies might like to stick a picture of a Terminator or Marty McFly style hoverboard in their advertising, that is very much an exaggeration and we are a few decades away (at least) from that kind of fully autonomous world.

AI is about data. The most common application is known as machine learning. This technique trains computer algorithms to think like humans do using existing data from the world around it. For example, we could load a computer with thousands of pictures of cats and dogs. The computer will ingest those images, convert them to data and going forwards, will work out for itself whether an animal is a cat or a dog.

To put this into a real-world context. An IoT device is anything that connects to the internet. Most people have countless technologies that do this from smartphones to laptops, tablets, Alexa, Google Home, televisions and even kettles, fridges and washing machines. When we speak to Alexa (the IoT device), it takes what we say and converts that into data using a method called natural language processing (NLP). That data is then matched against a huge database of previous conversations to find the closest match. Alexa can then reply to your command. Every time you have a conversation, Alexa learns and becomes more accurate.

So, for every IoT device we have, the more it gets used, the more useful it becomes as they learn from data. This is one reason that we are seeing a movement towards Industry 4.0.

What is Industry 4.0?

When we talk about anything “4.0” right now, it generally refers to the next evolution of technology. In this case it refers to industry and manufacturing. The advances in computer power, data, AI and IoT will orchestrate rapid advancement over the coming years and by the end of the decade, it is fully expected that the world will look quite different.

We are at a stage now where there is enough computing power and data available in the world to make fast progress within industrial settings, hence the buzz around Industry 4.0.

Whilst in the third industrial revolution (Industry 3.0), computers started being introduced, in this forward thinking era, having connected machines that communicate with each other is key without the need for human involvement. This will be a combination of physical systems, IoT and AI to create a data driven smart factory that is more efficient, productive and cost effective than we have ever seen before.

The network of all these machines is what we are calling Industry 4.0.

Key applications of Industry 4.0

Before we look at how you can set about driving an Industry 4.0 data strategy, here are a few of the use cases being utilised in practice.

Opportunities – with countless devices and sensors in a smart factory comes a vast amount of data. This can include information about the devices, environment, staff or outputs. It might take a human days to sift through millions of data points but new technology allows real-time analysis. For example, a sensor might flag that temperatures are too warm for production and reducing those will increase yield.    

Supply chain – data can be connected throughout the entire supply chain, from factory to the shelf for example. One common use case is in shipping when products are delayed and everything else connected to the system can be proactively adjusted.

Autonomous equipment – Most of us will be aware that autonomous vehicles are on the horizon with the likes of Tesla suggesting that driver-less cars could be on the roads as early as 2020 (there are already some being used in trials). In the Industry 4.0 context, the same technology can be used with cranes or trucks to make operations more efficient. For example, a crane that is controlled remotely rather than requiring a human driver.

Robotics – this is quite key to the future of Industry 4.0 as the machinery becomes more cost effective in factories. Amazon are a leader in the use of robotics, using such technology to move items around the warehouse, saving on resource cost and physical floor space.

To make all these things happen, factories or industries need to invest in the right kind of architecture.

Architectural Framework

Industry 4.0 relies on a large volume of devices and masses of data to be successful. Businesses must process data into timely and valuable information if they are to integrate it into production processes. For example, there is little use in having sensors that can track real-time maintenance issues if the infrastructure is not available to suitably analyze and act on those.

Unlike typical business data, IoT devices and sensors produce non-standardized and unstructured data. This is information that does not have a pre-defined model, usually very text heavy but could contain numbers and dates as well. Big Data Analytics platforms are required to manage the complex algorithms and programming models. It is one reason why we haven’t seen huge adoption to this point as industries don’t have the expertise in data science required to progress.

An Industry 4.0 architecture will vary depending on your business but in the main, the features below will be the key considerations for IoT and Data.

• Gateway – sensors need to connect seamlessly throughout the framework. Having equipment that talks to each other is imperative for communicating data at the right time. This will usually require ethernet cables or wireless mechanisms, both of which have reducing costs making it feasible for factories to install
• Edge computing – these are router services that can make fast, low latency decisions allowing for real-time data analytics. Edge services tend to sit nearer to sensors and machines to do faster communications, they do not connect to a wider Data Lake (see below)
• Ingesting data – multiple data sources need to be transformed into standard formats so they can be used in decision making processes. Data professionals with experience in such transformations will be able to apply the appropriate architecture for the best speed and consistency
• Data Lake – if you have all of this data, it needs to be stored somewhere. A Data Lake is a cloud based platform like Amazon Web Services (AWS) or Microsoft Azure or Google Cloud Platform (GCP) that is entirely scalable to the business needs and can be accessed from anywhere. Various scripting languages and libraries can be added to Data Lakes depending on the data ingestion strategy.
• Analytics and ML - understanding patterns and developing accurate models will require good quality data at scale and will lead to significant gains in overall productivity.
• Data Visualization – data from the Data Lake needs to be presented in such a way that the business can use it. Many business will do this using API’s that connect to commercial platforms like Tableau, Qlik, PowerBI or MS Dynamics to name a few.
• Data Security – appropriate software must be added to devices to ensure they are secure with the vast amounts of data they are producing and transforming.
• Automation of Data Flows - Robotics process automation (RPA) platforms such as UiPath or Automation Anywhere or Blueprism would provide BOTS framework and API’s to delegate system workflow autonomously without any human intervention.

Each of the elements above will have more detailed specifications but each of them are vital in having the right framework for Industry 4.0.

Top 10 Steps to Achieve Your Goals

Having the technology and data alone is not enough to succeed. Businesses need a well thought out strategy that clearly defines the objectives for Industry 4.0. Here are ten things that everyone should be thinking about.

1.      Have a clear objective

It is important to start with clear objectives for your Industry 4.0 data strategy. Evaluate exactly how mature you are at the start and what it will take to close the gap. Work backwards from where you want to be. Technology can be expensive so by prioritizing the measures that bring most value to your business, budgets can be correctly aligned.

2.      Senior executive support

Industry 4.0, data and AI is not about a team of Data Scientists coming up with solutions. It needs support from the top down. As soon as a business begins to automate processes, job roles changes and people need to find a way of accepting the transforming business culture. A top-down approach helps to remove any potential blockers

3.      Experimentation

Industry 4.0 relies on having massive amounts of data. On day one, you might be starting with a blank canvas. For example, how do you know the optimal temperature for production if the sensor has only just been installed and not created data yet? Small experiments will start gathering data and learn over time. You might start with a narrow scope and widen that with experience.

4.      Be pragmatic

If standards and infrastructure don’t exist yet you are not going to be able to set those up overnight. Be pragmatic in your approach and aim for the “low hanging fruit” before starting to climb Everest.

5.      Analyze the gaps

The chances are that you need a lot of support to achieve your objective. The right technology is difficult to develop and the right skills can be hard to find. It is important to fully review the market, finding the right people to work with. Include strategies for how you will develop your own employees to fit the new infrastructure.

6.      Data expertise

Industry 4.0 depends on data and analyzing it in creative ways to spot opportunities or potential efficiencies. Businesses need to learn how to get the most out of the mass data coming from their devices and use it to make decisions. Again, starting small with proof of concept ideas is key and using the data to support any claims you have.

7.      Let’s get digital

The whole business will need to adapt to a digital culture. You may have people who have worked in a factory for a long time and feel more comfortable working with pen and paper. This simply won’t be feasible in an Industry 4.0 environment and you will need to find ways that help them develop.

8.      Change is incremental

Industry 4.0 is not a business project that has an end date. There must be a continuous cycle of improvement as new data and opportunities are discovered. If anything, companies should strive to keep getting faster and making best use of new technology to stay ahead of the game.

9.      Collaboration is king

Working with third parties can be a useful exercise, especially in smaller businesses who do not have access to large volumes of data. Sharing of knowledge can be beneficial for all parties where they do not have conflicting interests. The success of Industry 4.0 relies on working with other digital leaders.

10.  Data, Data, Data

Ok, we’ve spoken about data a lot in relation to Industry 4.0 but it is almost impossible to stress just how important having adequate capabilities for data collection and storage is to success. Using algorithms to process data and spot outliers is imperative whilst also providing traceability of processes that are a cornerstone of Industry 4.0. Some have said that data is now a more valuable asset than oil and Industry 4.0 certainly has the potential to realize this.

Summary

Industry 4.0 is here and if businesses are going to remain competitive, they need to invest else risk falling quickly behind. AI, IoT and data are starting to truly power the future and those making best use of new technology and information are already seeing improved efficiency, productivity and cost reduction. The AI revolution is in full swing.

You can reach out to us on [email protected] for any help in this area.

 --------------------------------------------------------------------------------------------------------

Disclaimer: This publication contains general information and is not intended to be comprehensive nor to provide professional advice or services. This publication is not a substitute for such professional advice or services, and it should not be acted on or relied upon or used as a basis for any investment or other decision or action that may affect you or your business. Before taking any such decision you should consult a suitably qualified professional advisor. While reasonable effort has been made to ensure the accuracy of the information contained in this publication, this cannot be guaranteed, and neither associated organization nor any affiliate thereof or other related entity shall have any liability to any person or entity which relies on the information contained in this publication. Any such reliance is solely at the user’s risk. This article may contain references to other information sources.