Digital Transformation to Reduce Human Errors

28 April is the SafeDay - the World Day for Safety and Health at Work by the International Labour Organization (ILO) so like in previous years I’d like to bring attention to opportunities to improve safety in plants through digital transformation by extending the digital operational infrastructure (DOI) as part of Industrie 4.0 initiatives for operational excellence.

Imagine a plant where routine data collection is automated to avoid action slip like reading a gauge wrongly. And with automated checks on routine tasks to reduce incidents due to memory lapse like forgetting to close a valve. Or a more immersive learning experience to not apply the wrong rule due to inexperience. Geofencing to discourage violations like taking a shortcut through a high-risk area. As well as audit trail to discourage not following procedures. Reducing such human errors make the plant a safer place to work. Many plants have already started implementing these new best practices. So what are the real improvement opportunities? Here are my personal thoughts:

Manual tasks and human errors

Although plants have automatic control systems, there are still many manual tasks carried out in plants. Whenever there are manual tasks there is a possibility of human error; the human factor. Human errors are broadly classified in the skills-rules-knowledge framework illustrated above.

Skill-based errors are those when a skilled person does something wrong in the execution of the task. This includes action slip in routine tasks that requires little conscious thought for instance reading a gauge wrongly, reading the wrong gauge, transposing digits when writing down values, opening the wrong manual valve, or opening instead of closing etc. Skill-based errors also includes memory lapse such as forgetting to carry out a step of a task for instance forgetting to close a manual valve. 61% of human errors are skill-based.

Mistakes are those errors in the though process for planning a task due to inexperience or lack of information.

Rule-based errors are those mistakes where rules are in place. This includes applying the wrong rule, that is following the wrong procedure for instance due to inexperience or lack of information. Another rule-based error is when you follow a bad rule, that is there are errors in procedure, or the procedure is not up to date. Rule-based errors also include violation, that is intentionally not following the procedure, for instance to simplify or speed up the task such as taking a shortcut through a high-risk area of the plant or performing lifting operations even though supervisor is not present. This does not include sabotage which is not a human error. 27% of human errors are rule-based.

Knowledge-based errors are those mistakes where no rules are in place, that is, no procedure, such as in a new or unexpected situation where the person therefore has insufficient knowledge. 11% of human errors are knowledge-based.

Digital transformation of work practices

Many plants are now looking to digitization to reduce human errors to improve safety through new digital ways of working.

Automatic data collection

By automating previously manual data collection, plants can avoid action slip errors like reading a gauge wrongly, for instance due to parallax error, looking at the wrong gauge, or transposing the digits when writing down the reading by using wireless sensors instead of field operators with clipboard, pen, and paper doing the data collection manually. Actions on this data can now also be automated. With that, actions are taken based on correct readings resulting in fewer incidents due to decisions taken on false information. Plants can also reduce wrong rule errors like selecting the wrong procedure for the task due to lack of information by using innovative sensors to collect the data personnel need to take the right decisions, resulting in further reduction in incidents – and also reducing the burden on personnel. This is increasingly common in modern plants.

Automated checks on routine task

By automating checks on routine tasks for positive confirmation, plants can reduce incidents due to memory lapse errors like forgetting to close a manual valve, turning the wrong valve, or not closing it tight, by using valve position monitoring with interlocks instead of manual tracking of valve lineup resulting in fewer incidents due to incorrect manual operations on bypass, isolation, and transfer valves and other equipment – and also reducing the burden for confirmation and checks.

Audit trail

By automatically logging manual operations like opening and closing manual valves in an audit trail by using manual valve position monitoring, procedure violation errors in these operations are documented. With that, knowingly skipping steps, or performing steps like opening and closing valves in another sequence is discouraged resulting in reduced incidents due to people not carrying out all the steps in the procedures or doing it in a different order. The audit trail also helps in post incident forensic investigation, so personnel can learn from mistakes to avoid future repetition resulting in further reduction in incidents.

Automatic Identification (Auto-ID)

By automatically identifying equipment like valves, plants can reduce action slip errors like working on the wrong equipment by using RFID tags instead of reading name plates or tag plates that may be painted over, chipped off, scratch, or otherwise unreadable. This positive identification results in fewer incidents due to working on the wrong equipment.

Geo-fencing

By virtually ‘fencing up’ areas of the site, plants can reduce violation errors like taking a shortcut through a high-risk area of the plant or lifting operations without a supervisor present by using location awareness to report intruding, people are discouraged from straying into unauthorized areas resulting in reduced injuries due to hazards that may be present such as toxic gas. If personnel is working in the wrong area, this is easily detected. This is over and above other location awareness safety benefits such as automatic mustering headcount, rescue locating, man-down, and fatigue management. This is particularly valuable during shutdowns/turnarounds/outages when thousands of people unfamiliar with the plant are deployed at site. Learn more about location-based solutions and more from this digital transformation video.

Visual guidance

By visually guiding field operators, plants can reduce action slip errors like closing the wrong manual valve and wrong rule errors like selecting the wrong procedure for the task by using Augmented Reality (AR) to guide personnel to the correct piece of equipment to work on and display the information they need to pick the right procedure for the task resulting in fewer incidents due to working on the wrong equipment or following the wrong procedure. In addition to giving directions to assets and displaying equipment information and status, AR displays work instructions and asset connections. Use of nw technology like AR makes the site more attractive to the new generation of engineers.

Immersive learning

By making the learning experience more immersive, plants can reduce wrong rule errors such as selecting the wrong procedure for the job due to inexperience by complementing traditional training with Virtual Reality (VR) highly realistic classroom training instead of only on the job training. VR enables practice and learning in the safety of a classroom. You can make mistakes without consequences, then try again until you get it right every time. With that, personnel are better trained to take the right action in the field resulting in reduced incidents due to knowledge gaps and unfamiliarity with new situations. Use of technologies like VR makes the plant more attractive to young engineers.

Expert collaboration

By enabling expert collaboration, plants can reduce action slip errors like opening valves in the wrong order or at the wrong time by allowing an expert in a central location to see what the technicians looks at in the field through a live two-way video conference connection to give step-by-step instructions instead of attempting to guide over the phone or via email. Correct manual operations result in fewer incidents due to performing steps in the wrong sequence like isolate-equalize-vent etc. This solution is also ideal for collaboration between field operators and control room operators. This is over and above saving problem resolution time and travel costs.

The most interesting fact is that plants can use these digital solutions to meet the requirements of the safety case regime.

Deploy Digital Operational Infrastructure (DOI)

To enable digital transformation to new digital ways of working to reduce human errors, plants are now deploying Digital Operational Infrastructure (DOI) as a second layer of automation on the side of the basic process control, functional safety, and historian. The skill and knack of I&C engineers is “instrumental” (pun intended) to digital transformation.

Tens of thousands of plants have already started digital transformation by deploying WirelessHART infrastructure for automated data collection, replacing mechanical pressure and temperature gauges, sight level glasses, variable area flow meters, and dip sticks etc. with wireless sensors. Plants built on FOUNDATION fieldbus can also easily add more sensors at a marginal total cost by simply connecting those to the nearest junction box with a short piece of cable as the buses often have 100% spare capacity.

Manual valves are retrofitted with position sensors for position feedback to be used for the interlocks in control system and for the audit trail.

Valves and other equipment are retrofitted with RFID tags, and field personnel equipped with RFID reader and software.

Several plants with plant-wide Wi-Fi infrastructure have already deployed industrial grade (hazardous area approved) Real-Time Locating System (RTLS) and personnel have been issued hazardous area approved location tags. This technology is developing rapidly right now. New location awareness solutions instead use WirelessHART infrastructure, with more cost-effective location anchors and easily rechargeable location tags.

Augmented Reality (AR) technology is in its infancy, particularly industrial grade (hazardous area approved) hardware. However, the backend software and mobile device-end apps is already very capable and easy to use. Plants are deploying plant-wide Wi-Fi infrastructure and providing cloud connection for multiple applications, including to be prepared for AR.

Virtual Reality (VR) solutions are already mature. They can operate stand-alone or integrate with modeling and simulation (“Digital Twin”) software and Operator Training Simulators (OTS) using standard OPC-UA.

Wearable cameras, essentially an industrial (hazardous area) tablet you wear on your head, are used for hands-free video conference with an expert in another location. Sites that have plant-wide Wi-Fi infrastructure and internet connection can use this solution.

The future is digital – this is what digitalization and Industry 4.0 is all about. The plants most successful with digital transformation start small, scale fast with readymade solutions in a phased approach, but think big beyond safety, including reliability, energy efficiency, and productivity. The transformation journey starts with a discovery workshop from which a roadmap can be created. Well, that’s my personal opinion. If you are interested in digital transformation in the process industries click “Follow” by my photo to not miss future updates. Click “Like” if you found this useful to you and “Share” it with others if you think it would be useful to them.