Top 5 ERP News and articles: Week 2019-7: Industry 4.0

Disclaimer: The articles in this blog post are those that I found interesting and relevant to the topic of ERP and broader technology. I have no commercial association with any of the entities mentioned in this article. I may be subscribing to a few news letters,  following a few of these entities on LinkedIn and even some of these entities may be on my LinkedIn or Social Media network. These articles are selected purely based on their relevance to the objective of this blog, to promote ERP.

Theme for this week is Industry 4.0 (I40). The articles selected equips you to better understand the complex area of I40. Three of the articles are conceptual in nature and are meant to provide overview. The first article talks about I40 (It is a post from this blog), second gives an overview of IoT and third gives an overview of AI. Fourth article details a few use cases of I40 Technologies

The Short URL for this Post is  https://goo.gl/jNWE7g

1. Industry 4.0: A Primer:

 http://erp-consultancy.blogspot.com/2018/03/what-is-industry-40.html#more

Industry 4.0 (I40) refers to the fourth stage in the evolution of manufacturing industry. Phase one was characterized by Steam Energy, two by electricity, three by computerization and IT and Industry 4.0 envisages the integration of cyber physical systems, big data, modern technologies - AI, ML and IIoT and cloud computing that is set to transform the industry.

I40 is driven by four technology disruptions, rise in data volume and computational power, augmented analytic capabilities, new forms of human - machine interaction like touch and AVR (Augmented Virtual Reality) and digital to physical integration like robotics and 3D Printing.
I40 factory of the future will be Self aware, will self-predict, self-compare, self-configure, self-maintain and self-organize.

'Customerization', producing a batch size of one, is the major benefit of I40. Customer will truly become the king. Customer will be able to drag and drop features that they require and the system will produce plan for and produce it to meet the specific requirements of individual customer.
The detailed article (from my blog) shows the challenges and opportunities of I40 and closes with a brief description of how an ERP system can become I40 ready.
There are many references at the end of the article that can add more value to the reader.

2. Getting started with Internet of Things (IoT) 

https://www.sas.com/en_us/explore/resources/getting-started-iot.html?utm_source=linkedin&utm_medium=cpc&utm_campaign=iot-global&utm_content=GMS-81562

 I love it when I open a website and find that the contents are laid out in a logical order and in a design that is pleasing to the eyes. Scanning the website, one gets a mental sitemap of what it is about. It makes me very excited to explore the site. The website on IoT maintained by sas.com has material to learn everything about IoT. Their structured approach to IoT consists of five conceptual steps,  identify challenges, determine goals, put your goals into action by developing a monetization plan, engage the ecosystem and reflect on your IoT journey and take the next steps. For each of these steps, the site has provide with collateral like articles, whitepapers, podcasts and videos. 

I will summarize the article on 10 common mistakes organizations make with IoT endeavours. The ten mistakes are:

• Lack of preparation - preparation will take time and money, 
• Inaccurate problem statement - this leads to excessive and redundant data collection and the mistake of treating all data as equally important (some data is more important), 
• Not asking the right questions - data ownership, data quality, how to handle adhoc analytics etc, 
• Failing to anticipate the complexity - not anticipating data volume and not planning for scalability, 
• Tight restriction on data exchange between technologies and applications, 
• Skipping compliance best practices - compliance processes like GDPR and others detail best practices for collecting, storing and disseminating data. They should be adhered to, 
• Underestimating security and privacy implications - IoT demands additional security and privacy demands. When everything is connected, everything is at risk, 
• Taking data platforms and people skills for granted - take a hard look at your infrastructure, company culture and policy and employee skills and see if they are sufficient to meet the demands of IoT., 
• Complacency - do not assume that the product vendors and consultants have all the answers, they come with their own agenda that may not match yours
• Waiting until every relevant information is available - get involved, execute, learn and improve

Excellent article. Excellent resources.

3. An executive's guide to AI  

https://www.mckinsey.com/business-functions/mckinsey-analytics/our-insights/an-executives-guide-to-ai

This is an awesome, visual treat of a site that gives you a very solid grounding on the key concepts relating to AI - AI itself, Machine Learning (ML) and Deep Learning (DL). Delivered by McKinsey Analytics team, the site is divided into three sections. Each section covers definition and additional details. This site also describes detailed use cases of both ML and DL. Under AI, it covers definition and a timeline that tells you why AI is in the vogue now. The article identifies algorithmic advances, high volume of data and increase in computational power as the drivers for AI. Algorithmic advances started in 1805 when Legendre published the least square method for regression which laid the foundations of Machine Learning. In 2010 the world wide IP traffic exceeded 20 exabytes and in 2011 IBM Watson won Jeopardy!

Machine Learning provides predictions and prescriptions. Major types of ML are Supervised Learning, Unsupervised Learning and Reinforcement Learning. Each of them is defined briefly. For example RL is defined where 'Algorithm learns to perform a task simply by trying to maximize rewards it receives for its actions'.

Deep learning can process a wider range of data resources, requires less data processing by humans and can produce more accurate results than traditional ML. It uses 'Neural Network' that can process vast amounts of input data through multiple layers that learn increasingly complex features of the data at each layer. There are two major DL models, Convolutional Neural Network and Recurrent Neural Networks

There are many use cases discussed for each type of ML and each model of DL.

If you are interested in AI, you cannot afford to miss this site. Thank you McKinsey for this awesome site.  

4. AI in ERP: Quest for the fourth generation platform  

https://it.toolbox.com/blogs/erpdesk/ai-in-erp-quest-for-the-fourth-generation-platform-073118

Industry 4.0 expects new advanced technologies to work with the existing technologies to automate manufacturing processes. AI is one of the new technologies that are expected to work seamlessly with ERP to add value to the repository. In fact AI is going to have transformation impact on ERP Development and could lead to 'Post Modern' ERP, or Fourth Generation ERP. Some of the use cases of AI include the use of Chatbots and Virtual Assistants that work with ERP releasing time for employees to do value added work on ERP. Advanced Analytics using AI will add 'meat' to the 'data skeleton' provided by ERP. In the area of UX, Another use case for AI is that it helps add 'Voice Interaction' capabilities to ERP. 

5. 10 Trends shaping the IoT Landscape: McKinsey.com

https://www.mckinsey.com/business-functions/digital-mckinsey/our-insights/ten-trends-shaping-the-internet-of-things-business-landscape?cid=other-eml-alt-mip-mck&hlkid=4594adde5e744a7a983d6c9f94a92f27&hctky=2565821&hdpid=1117bc77-3cf1-44f3-a5b5-6d0292addfc0

The article places the IoT trends into three buckets, Company Level, Market Level and at the Technology / data level.

Companies no longer views IoT as an IT initiative. They have come to realize that IoT and business processes are closely interrelated. For example, by attaching sensors, a machine will inform you of the pending predictive maintenance. However, that output is useless unless it is linked to maintenance planning and spares procurement. Also, companies gain the maximum value from their IoT investment when they execute multiple use cases. To add to above, IoT is slowly enabling more subscription based business like 'compressed air by the hour', but consumers, especially at households are still resistant to the subscription model.

At the market level, heavy machinery and discrete manufacturing is slowly extending the IoT usage to bring about significant cost benefits. In addition, Amazon and Google have attained critical mass in connected devices. The upshot will be that manufacturers will position their products to work with these appliances. The emergence of Chinese IoT companies is another trend that will act as a headwind to the overall IoT industry worldwide.

At the Technology and Data level, the authors of the article expect four trends. Data integration between connected devices is still a challenge and businesses are not yet able to get the full value of IoT. The emerging scenarios in data sharing will depend on industry. Some industries like aircraft will encourage data sharing while some industries would like to keep tight ownership of data to differentiate their performance. Another trend will be the closure of debate relating to whether cloud of 'Edge' is the best option to host IoT data. 'Edge' means servers near to the place where data is generated. The decision will depend on data transmission costs and if the company has a number of mobile and / or remote assets. If the transmission costs are high and company has many remote assets, saving IoT data on Edge will help faster analytics. Another trend is that 'Cybersecurity' is not a barrier to IoT adoption. The final trend is that AI and ML are catching up with IoT and are capable of doing about 60% of the tasks that IoT does. This could lead to a situation where they replace IoT in the investment priority.

Nice article.....Do read. Do subscribe to McKinsey, if you have not done already.  

Additional Reading

1. Five ways in which CIOs are deploying AI 
2. What is Industry 4.0: Bernard Marr on Forbes.com 

3. The future of manufacturing 
4. Factory of future to be showcased at ATX West 2019