An IIoT SWOT analysis

16 March 2020

José Miguel Dias Pereira summarises the main strengths, weaknesses, opportunities and threats of the Industrial Internet of Things (IIoT) and offers an overview of the real-time demands of industrial networks, at plant level, at DCS level, and at SCADA level.

Every new step in any evolution process will always some strengths and weaknesses. This is particularly true in IIoT where there is an integration of different equipment that works in the same or different networks, the same or different locations, and services that are included in the value chain of a given manufacturing product. 

While legacy industrial networks were limited in terms of their data acquisition capabilities. Today, when large amounts of data is transmitted through the Internet, new threads emerge – such as security. The use of the IP security (IPSec) with the IPv6 site-to-site tunnel mode capability, reinforces and extends the security mechanisms provided by VPNs – the encryption and integrity checking of VPNs is always implemented in all IPv6 connections.

The implementation of the IIoT stimulates the creation of new business models based on the large amount of data it creates. Data analytics also enables a huge increase in system reliability and performance by analysing operational data of industrial equipment that supports condition-based maintenance, predictive maintenance, instrument asset management and returns on asset management gains. The use of powerful predictive data models makes it possible to extract information that traditionally would be lost. However, in each IIoT application there are specific issues that requires a multidisciplinary expertise, in terms of engineering, data processing and computing, to analyse, consolidate and extract relevant information. 

While big data can provide insight on industrial processes, the collection huge amounts of data without criteria is not a good procedure. For example, if an industrial system is upgraded and new instruments, with enhanced characteristics are introduced, it makes no sense to perform data analysis mixing data collected before and after the upgrade.
On the other hand, it is important that old data is stored because it does contain information that can be essential to evaluate the process gains that are associated with the upgrade of a given industrial system. 

Industrial networks
IIoT is supported by networks and protocols which support the industrial communications between different industrial devices and the communication of industrial platforms to services, maintenance centers, logistics and data processing centers, for example. So, networking and protocol compatibility is essential for the success of IIoT. The success of any evolution process depends on the past and present, so it also makes sense to present a brief historical review of the main communication solutions used in manufacturing units. It is important to remember that the first step towards standardisation of communication between different industrial devices has, traditionally, been based on the usage of normalised pneumatic and electrical signals. A large number of industrial networks are still based on current loops that use a current amplitude variable between 4 mA and 20mA to translate the value of the variable under measurement. Usually, there is a linear relationship between the variables, being the minimum current value (4 mA) associated with the lower range value (LRV) of the variable under measurement and the maximum current value (20 mA) associated with the upper range value (URV) of the variable under measurement.

Even if there are a large number of limitations associated with current loop transmission, it is important to note that a large number of manufacturing units are still using this signalling mode as the period of time for return on investments (ROI) in old manufacturing units can be higher than a few tens of years within acceptable production performance levels. With the advent of digitalisation and advanced instrument capabilities a hybrid analogue and digital communication mode was introduced. In addition to the 4-20 mA of the current loop signal, a superimposed Frequency-shift keying (FSK) digital signal is used to transmit digital data to and from instrument devices. The large number of HART commands that include universal, common practice and device specific commands, makes it possible to take advantage of the new capabilities of smart sensing instruments. 

A third step of industrial network and protocol evolution is associated with the use of digital communications with higher bit rates than those provided by HART – For example, Foundation Fieldbus and Profibus – some of these protocols can share deterministic and random traffic in the same communication support. Deterministic traffic is assigned to specific time slots of the data frame which occur periodically. The remaining bandwidth of the communication support is used for unscheduled communication that occurs randomly. These networks enable the implementation of field control solutions (FCS)  – the acquisition, control and data processing tasks implemented in the field devices – with no processing overload of the DCS. 

A fourth step, in the industrial networks and protocols evolution relates to  the introduction of industrial Ethernet. Several proprietary solutions were introduced with increased performance in terms of latency and jitter but those solutions are not open so gateways are required to connect different networks. Several IEEE standards are now being developed for time sensitive networking (TSN) solutions. These standards [3-9] promote the implementation of real time capabilities in Ethernet networks and devices interoperability without the need for special interfaces and gateways.

A SWOT analysis of the IIoT
It makes sense to perform a technical SWOT analysis about the IIoT because its success depends on many factors. This analysis can be a powerful tool to help develop industrial business strategy

Strengths: One of the main strengths of IIoT is its capability to connect a huge number of devices through the Internet using wired and wirelessly connection modes. These devices can include simple sensors or actuators, machine, cloud processing platforms and different capabilities, such as machine to machine (M2M) communication, integrated asset management systems (IAMS), data analytics and storage platforms, and logistic services. The ubiquitous characteristic of the Internet enables communication between multiple industrial devices that can be working in different manufacturing units (MU) to different services that are included in the value chain of a manufactured product. For example, it is possible to use the sales units of a particular product to schedule the production level of that product in the MU. Another example relates to the reliability of instrumentation within a system – using data from sensing units it is possible to implement an integrated predictive maintenance of systems.

In addition to cost reductions, innovations and new services can also be fostered through the IIoT thanks to the huge amount of data created. As long as adequate data is gathered and wise analysis techniques are used, it is possible to make better decisions and to extract important information from data that was considered as uncorrelated. The results from data analysis can improve the visibility into system behaviour, helping increase  the competitive advantage of a MU.

Weaknesses: Security, data processing challenges, and cost are the main weakness associated with IIoT. Security will always be the weak point of any system or service connected through the Internet. While security and encryption techniques are continually being improved, hacker activities never stop and the connection of a huge numbers of devices to the Internet increases their chances of a successful attack. 

The firmware that runs in low-cost IIoT devices is vulnerable to attack. Distributed denial-of-service (DDoS) attacks are employed by hackers to overwhelm an Internet client or server, preventing Internet traffic from accessing network nodes. 

It is also important to note that an excess of outdated data offers no added value and can be an IIoT weakness. Data collection, storage and processing should judiciously use the limited resources of a computing system and this not an exception in cloud computing platforms. The cost of connecting a huge amount of industrial devices and systems is also an IIoT weakness. Costs are not limited to tangible costs such as new equipment and infrastructure. There will also intangible costs associated with the research and development for new products and software. 

The development of data analytical tools is usually a complex task that involves many partners because the best solution with be bespoke for a given application. So achieving a successful IIoT implementation requires the investors to be patient about process improvements and profit returns because they will not come instantly. 

Another IIoT weakness relates to its novelty – a lack of standardisation and compatibility risks. There is no standard of compatibility that assures a non-risky business scenario to investors.

Opportunities: There are several opportunities associated with the IIoT. Most important is the opportunity to replace old MUs with new MUs to achieve higher productivity yields through greater process insight from analysed data. New services, supported by increased connectivity, are also possible. For example, in the maintenance area, giving remote access to process data to external teams can improve the reliability of a manufacturing process by avoiding potential conditions that would cause production shutdowns or reduce the quality of manufactured products. Remote access to data allows for more successful implementation of predictive maintenance enabling  better analysis of equipment working conditions and avoiding failure risks. 

Other opportunities come from improved production flow monitoring – in a single MU or in a set of MUs that work collaboratively; greater control of inventory management helps avoid an excess or shortage of raw materials; improvements in plant safety and security come from analysing data associated with hazardous events; and quality control improvements can be achieved by processing data that affects the quality parameters of a given product.

Threats: The main IIoT threat relates to a greater vulnerability to cyber attacks. As the number of devices that will connect to the Internet increases  the negative impact of potential security failures rises. 

Another threat relates to the expectation of receiving a fast return on investment. Data analytic tools do offer the possibility to extract incredibly useful information from the MUs. However, if the data is not analysed correctly the right information may not be available to improve MUs performance. At the same time, new skills will be required to successfully  implement IIoT projects because operational technologies (OT) and information technologies (IF) functions will be fused. Multidisciplinary teams will need to work together to achieve the desired goals.

A final threat comes from the actual addressing mode used on the Internet. IIoT requires the adoption of IPv6 in order to extend the addressing capabilities of the networks, and to improve network security, connectability and scalability. However, the transition from IPv4 to IPv6 is being delayed because significant investment is required and there are also potential risks in the transition phase.

Conclusion
IIoT is not a panacea for all industrial network related problems. Indeed, at the plant floor level there are challenges that have no easy solutio, particularly when remote access to industrial data poses confidentiality, security and safety risks. There are also costs associated with the introduction of IIoT and returns of investment are not usually achieved in the short term. 

Big data analysis is a multidisciplinary task that requires expertise from engineering, data processing, computing and data analytics fields. As a final conclusion of this overview about IIoT, it must be underlined that its implementation is a medium-term process, it requires a specific tuning of each process and industrial plant, and that security issues must be properly addressed to give confidence for investments in this new industrial paradigm.

José Miguel Dias Pereira is principal coordinating professor at the Polytechnic Institute of Setúbal in Portugal.


Print this page | E-mail this page