Advanced automation and insight with Industry 4.0

BLOG

Reflections on Industry 4.0

by JOHN TAYLOR 31 Oct, 2017

Collecting and analysing additional data to provide an enhanced picture of the performance of an asset is the subject of this blog post.

First let’s consider the issue of building energy consumption and specifically, natural gas use for space heating. To determine whether more or less gas was consumed than the previous period and/or whether energy efficiency projects were effective, it is necessary to monitor more than just the amount of gas that was consumed.

Natural gas consumption for space heating use varies with ambient temperature. More gas is usually consumed in winter months than in summer and more gas is burned in very cold winters than in warmer winters. So to establish a more precise picture of actual building performance it is necessary to also capture ambient temperature data. This is usually done by collecting degree day data which provides a discrete measure of how cold or warm it was in a particular location on a particular day, (week or month). With this additional data it is then possible to analyse specific energy consumption in kWh per degree day. Hence if I know how many degree days in a month I can predict the expected energy consumption and compare it with the actual to determine performance.

A similar method can be applied in an industrial context to determine specific energy consumption where energy consumption varies according to production activity e.g. widgets made, man-hours worked, etc.

Another example of where additional data capture can be used to provide a more precise digital twin, is that of vibration monitoring to determine the condition of rotating machinery. Vibration sensors (accelerometers) can be easily retrofitted to appropriate points on rotating equipment e.g. motors, compressors, gear boxes, etc. and used to detect the deterioration of components with the aim of preventing catastrophic failure.

More data, can provide an enhanced picture of past performance and used to predict future performance but the next challenge (and the harder piece) is turning this intelligence into action that results in improved performance. We’ve seen lots of examples of very clever (and expensive) data collection and reporting systems that are ignored because the intended users don’t understand what the system is telling them and how they need to respond.

Assistance – the topic of the next blog post.

by JOHN TAYLOR 24 Oct, 2017

From our perspective information transparency has two levels, the first is concerned with deriving an improved picture of an asset’s performance based upon existing data and the second is concerned with creating an enhanced picture based upon additional data and analysis. This blog post is concerned with the former whilst a second post will consider the latter.

In 2004 we met a new client (who by the way is still a client today) who described an issue that we have since heard repeated many times, ‘I can’t easily access the data I need to assess the current and historic performance of the asset’. The key terms in this statement we discovered were easily , data and performance .

Our initial contact with the new client was a European Operations Manager responsible for a fleet of remotely-located, unmanned industrial assets. These assets are located at end-user sites and they autonomously deliver a utility under the control of a local PLC and SCADA system.

It was possible (albeit with some difficulty) to remotely connect to the local SCADA system and extract sensor data, via a remote control application. This data then had to be imported into a spreadsheet and processed to select the required data, before the desired picture of performance could be provided. On occasion is was discovered that the desired data had not been retained by the control system.

It was some time however before our contact was able to persuade his colleagues to invest in the development of what became known as an Asset Performance Management (APM) system, partly, at the time, because there was scepticism about the value of such as system, frequently the reply was ‘…but we’ve already got the data in our control system.’

Over a couple of years we helped this client development an APM system that was reliable, cost effective to implement and highly beneficial. As a result of the improved transparency provided by the system the client was able to reduce the cost of operations and maintenance and increase the profitability of the business.

This APM system is still employed today. It provides very easy access to up-to-date, useful information (as opposed to just data) concerning the current, historic and expected performance of an asset and it alerts users when attention is required i.e. it provides information transparency.

by JOHN TAYLOR 22 Oct, 2017

In many situations the Internet of Things is concerned with connecting devices that were previously unconnected.

For example level sensors on an LPG tank previously indicated the tank level via a visually read gauge on the tank but increasingly mobile enabled ‘things’ read the tank level sensor and periodically transmit the reading to remote servers. The LPG distributor is then able to schedule deliveries to avoid runouts whilst minimising their distribution costs.

In an industrial context however, whilst there are still sensors that are unconnected (e.g. gauges on tanks, utility meters, etc.), many are already connected to PLCs and SCADA systems for control purposes but that’s often where the connectivity ends. In an industrial situation the IoT is also concerned with extending this connectivity to other systems which may be hosted remotely.

For example remotely located plant, operated locally by a PLC system, may be monitored and controlled from a central operations centre if the PLC systems are in turn connected via the IoT. For a given fleet of assets however it’s likely that these PLC systems will vary in age, make and model, hence the importance of interoperability.

Obviously one of the key concerns of this extended connectivity is security. Whilst it might be highly advantageous providing this connectivity from a commercial perspective, clearly the security risks and their mitigation need to be evaluated and planned with sufficient care.

Mitigating security risks associated with IoT implementations will be the subject of another blog.      

by JOHN TAYLOR 18 Oct, 2017
Interoperability is described as a key requirement of Industry 4.0 but what exactly is it, why does it matter and how is it achieved?

Interoperability is concerned with connectivity and communication between machines, devices, sensors and people which in its self is nothing new. For many years now, across many industries, sensors connected to programmable logic controllers (PLC), themselves connected to supervisory control and data acquisition systems (SCADA), have been communicating data between themselves and with people.

Many of these systems however were designed to perform specific tasks and extensibility was often not an important design requirement. Increasingly however people have seen opportunities in accessing the data used by these control systems for other purposes (data which is now referred to as Dark Data) but the problem was gaining access to this data in these 'closed' systems.

One of the technologies that has made this connectivity and communication much easier is OPC (Open Platform Communications). OPC UA was first released in 2008 and the current version 1.03 was released in Oct 2015.

OPC UA allows data to be exchanged easily and securely between different platforms from many different manufacturers. If you need access to data being captured by one system for use by another system, for another purpose, you can do this easily with an OPC UA server.

For example a PLC 'run signal' (true/false) from an existing control system, can be used to automatically communicate a machine downtime state to an Overall Equipment Effectiveness (OEE) reporting system, via an OPC UA server.

What if the machine being monitored however is in a different geographical location to another system that requires access to the machine data?

This is where the Internet of Things (IoT) has a part to play. Interoperability and the IoT is the subject of the next blog post, Interoperability - Part 2. 
by JOHN TAYLOR 12 Oct, 2017
Industry 4.0 (or more correctly Industrie 4.0 - the term was first coined in Germany) refers to the fourth industrial revolution. From a simplified technological perspective the first industrial revolution was driven by the steam engine, the second by electrical energy and the electric motor and the third by programmable devices and robotics. The fourth industrial revolution is being driven primarily by the Internet of Things (IoT).

An Industry 4.0 implementation is characterised by four main features, interoperability (connectivity and communication between devices and between people and devices), information transparency (a more precise and informative digital picture of real-world status and activity), assistance (providing operators with technical advice on action to be taken) and decentralised decision-making (the ability of cyber-physical systems to makes decisions and to perform autonomously where human intervention was previously required).

It is estimated that Industry 4.0 has the potential to increase productivity by up to 50%, so it's definitely worth further investigation and that's what this blog intends to do.




Reflections on Industry 4.0

by JOHN TAYLOR 31 Oct, 2017

Collecting and analysing additional data to provide an enhanced picture of the performance of an asset is the subject of this blog post.

First let’s consider the issue of building energy consumption and specifically, natural gas use for space heating. To determine whether more or less gas was consumed than the previous period and/or whether energy efficiency projects were effective, it is necessary to monitor more than just the amount of gas that was consumed.

Natural gas consumption for space heating use varies with ambient temperature. More gas is usually consumed in winter months than in summer and more gas is burned in very cold winters than in warmer winters. So to establish a more precise picture of actual building performance it is necessary to also capture ambient temperature data. This is usually done by collecting degree day data which provides a discrete measure of how cold or warm it was in a particular location on a particular day, (week or month). With this additional data it is then possible to analyse specific energy consumption in kWh per degree day. Hence if I know how many degree days in a month I can predict the expected energy consumption and compare it with the actual to determine performance.

A similar method can be applied in an industrial context to determine specific energy consumption where energy consumption varies according to production activity e.g. widgets made, man-hours worked, etc.

Another example of where additional data capture can be used to provide a more precise digital twin, is that of vibration monitoring to determine the condition of rotating machinery. Vibration sensors (accelerometers) can be easily retrofitted to appropriate points on rotating equipment e.g. motors, compressors, gear boxes, etc. and used to detect the deterioration of components with the aim of preventing catastrophic failure.

More data, can provide an enhanced picture of past performance and used to predict future performance but the next challenge (and the harder piece) is turning this intelligence into action that results in improved performance. We’ve seen lots of examples of very clever (and expensive) data collection and reporting systems that are ignored because the intended users don’t understand what the system is telling them and how they need to respond.

Assistance – the topic of the next blog post.

by JOHN TAYLOR 24 Oct, 2017

From our perspective information transparency has two levels, the first is concerned with deriving an improved picture of an asset’s performance based upon existing data and the second is concerned with creating an enhanced picture based upon additional data and analysis. This blog post is concerned with the former whilst a second post will consider the latter.

In 2004 we met a new client (who by the way is still a client today) who described an issue that we have since heard repeated many times, ‘I can’t easily access the data I need to assess the current and historic performance of the asset’. The key terms in this statement we discovered were easily , data and performance .

Our initial contact with the new client was a European Operations Manager responsible for a fleet of remotely-located, unmanned industrial assets. These assets are located at end-user sites and they autonomously deliver a utility under the control of a local PLC and SCADA system.

It was possible (albeit with some difficulty) to remotely connect to the local SCADA system and extract sensor data, via a remote control application. This data then had to be imported into a spreadsheet and processed to select the required data, before the desired picture of performance could be provided. On occasion is was discovered that the desired data had not been retained by the control system.

It was some time however before our contact was able to persuade his colleagues to invest in the development of what became known as an Asset Performance Management (APM) system, partly, at the time, because there was scepticism about the value of such as system, frequently the reply was ‘…but we’ve already got the data in our control system.’

Over a couple of years we helped this client development an APM system that was reliable, cost effective to implement and highly beneficial. As a result of the improved transparency provided by the system the client was able to reduce the cost of operations and maintenance and increase the profitability of the business.

This APM system is still employed today. It provides very easy access to up-to-date, useful information (as opposed to just data) concerning the current, historic and expected performance of an asset and it alerts users when attention is required i.e. it provides information transparency.

by JOHN TAYLOR 22 Oct, 2017

In many situations the Internet of Things is concerned with connecting devices that were previously unconnected.

For example level sensors on an LPG tank previously indicated the tank level via a visually read gauge on the tank but increasingly mobile enabled ‘things’ read the tank level sensor and periodically transmit the reading to remote servers. The LPG distributor is then able to schedule deliveries to avoid runouts whilst minimising their distribution costs.

In an industrial context however, whilst there are still sensors that are unconnected (e.g. gauges on tanks, utility meters, etc.), many are already connected to PLCs and SCADA systems for control purposes but that’s often where the connectivity ends. In an industrial situation the IoT is also concerned with extending this connectivity to other systems which may be hosted remotely.

For example remotely located plant, operated locally by a PLC system, may be monitored and controlled from a central operations centre if the PLC systems are in turn connected via the IoT. For a given fleet of assets however it’s likely that these PLC systems will vary in age, make and model, hence the importance of interoperability.

Obviously one of the key concerns of this extended connectivity is security. Whilst it might be highly advantageous providing this connectivity from a commercial perspective, clearly the security risks and their mitigation need to be evaluated and planned with sufficient care.

Mitigating security risks associated with IoT implementations will be the subject of another blog.      

by JOHN TAYLOR 18 Oct, 2017
Interoperability is described as a key requirement of Industry 4.0 but what exactly is it, why does it matter and how is it achieved?

Interoperability is concerned with connectivity and communication between machines, devices, sensors and people which in its self is nothing new. For many years now, across many industries, sensors connected to programmable logic controllers (PLC), themselves connected to supervisory control and data acquisition systems (SCADA), have been communicating data between themselves and with people.

Many of these systems however were designed to perform specific tasks and extensibility was often not an important design requirement. Increasingly however people have seen opportunities in accessing the data used by these control systems for other purposes (data which is now referred to as Dark Data) but the problem was gaining access to this data in these 'closed' systems.

One of the technologies that has made this connectivity and communication much easier is OPC (Open Platform Communications). OPC UA was first released in 2008 and the current version 1.03 was released in Oct 2015.

OPC UA allows data to be exchanged easily and securely between different platforms from many different manufacturers. If you need access to data being captured by one system for use by another system, for another purpose, you can do this easily with an OPC UA server.

For example a PLC 'run signal' (true/false) from an existing control system, can be used to automatically communicate a machine downtime state to an Overall Equipment Effectiveness (OEE) reporting system, via an OPC UA server.

What if the machine being monitored however is in a different geographical location to another system that requires access to the machine data?

This is where the Internet of Things (IoT) has a part to play. Interoperability and the IoT is the subject of the next blog post, Interoperability - Part 2. 
by JOHN TAYLOR 12 Oct, 2017
Industry 4.0 (or more correctly Industrie 4.0 - the term was first coined in Germany) refers to the fourth industrial revolution. From a simplified technological perspective the first industrial revolution was driven by the steam engine, the second by electrical energy and the electric motor and the third by programmable devices and robotics. The fourth industrial revolution is being driven primarily by the Internet of Things (IoT).

An Industry 4.0 implementation is characterised by four main features, interoperability (connectivity and communication between devices and between people and devices), information transparency (a more precise and informative digital picture of real-world status and activity), assistance (providing operators with technical advice on action to be taken) and decentralised decision-making (the ability of cyber-physical systems to makes decisions and to perform autonomously where human intervention was previously required).

It is estimated that Industry 4.0 has the potential to increase productivity by up to 50%, so it's definitely worth further investigation and that's what this blog intends to do.




Share by: