<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[HyBird - Medium]]></title>
        <description><![CDATA[Using CV and AI to Streamline Asset Management - Medium]]></description>
        <link>https://medium.com/hybird?source=rss----c70a352ce428---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>HyBird - Medium</title>
            <link>https://medium.com/hybird?source=rss----c70a352ce428---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 14:23:49 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/hybird" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Why your organisation should ditch P&IDs right now — Part 2]]></title>
            <link>https://medium.com/hybird/why-your-organisation-should-ditch-p-ids-right-now-part-2-74f0aaa3cf8b?source=rss----c70a352ce428---4</link>
            <guid isPermaLink="false">https://medium.com/p/74f0aaa3cf8b</guid>
            <category><![CDATA[iot]]></category>
            <category><![CDATA[oil-and-gas]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[computer-vision]]></category>
            <category><![CDATA[asset-management]]></category>
            <dc:creator><![CDATA[Mohamed Hafez]]></dc:creator>
            <pubDate>Mon, 01 Mar 2021 10:31:46 GMT</pubDate>
            <atom:updated>2021-03-01T10:31:46.380Z</atom:updated>
            <content:encoded><![CDATA[<h3>Why your organisation should ditch P&amp;IDs right now — Part 2</h3><p>In <a href="https://medium.com/p/d210d90e655b">Part 1</a>, it became obvious why conventional P&amp;IDs should be abolished — and that they should be replaced with Intelligent or Smart P&amp;IDs (S-P&amp;IDs) instead:</p><blockquote>Smart P&amp;IDs allow engineers to access all necessary site information from <strong>one, always up-to-date, digital hub</strong>.</blockquote><p>So let’s take a deeper look at how you can start using S-P&amp;IDs today…</p><h3>How do you even make S-P&amp;IDs?</h3><p>Implementing S-P&amp;IDs is typically tedious, resource-intensive, and expensive, with vendors completing tonnes of heavy-lifting upfront: from inventing P&amp;IDs to developing bespoke software and tools. Instead of starting from zero, it’s best to use AI to convert existing P&amp;IDs into S-P&amp;IDs.</p><h4>Consolidating Information</h4><p>First, identify all of the key site data that relates to asset management. This often includes:</p><ul><li><strong>Component Information</strong>, such as component details, IDs, manufacturers, materials, weights, etc.;</li><li><strong>Process Information</strong>, such as powers, pressures, flows, and much more; and</li><li><strong>Operational Information</strong>, such as work orders, notifications, reports, and more.</li></ul><p>To produce a true single-source-of-truth, it’s imperative that all of this information is integrated and accessible through one intuitive platform.</p><h4>Contextualising Information</h4><p>After identifying and locating all of this information, it needs to be tagged to the S-P&amp;ID. The 2D PDF/DWG serves as a background layer, allowing tags to be placed on top of the corresponding components. Once the components are tagged to the S-P&amp;ID, users can simply click on them to view comprehensive datasets.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hEVQP6uinaR-9Ld-Dw5mBA.png" /><figcaption>S-P&amp;ID displaying information relating to component “Pump P-3” (highlighted using green bounding box)</figcaption></figure><h3>But wait, there’s more!</h3><blockquote>“Computer Vision can make your life even easier.”</blockquote><blockquote>- Everyone.</blockquote><blockquote><strong>Computer Vision</strong> (also known as Machine Vision) is an interdisciplinary field of Artificial Intelligence concerned with providing computers, machines, and robots with the ability to understand digital images and videos.</blockquote><p>Computer Vision (CV) techniques have made the process of creating S-P&amp;IDs 80% faster, and 50% cheaper than alternative methods.</p><h4>Automating Inefficiencies</h4><p>The most time-intensive routine in creating S-P&amp;IDs is finding the correct P&amp;ID file for the component, followed by digitally drawing the precise bounding box, then tagging the relevant data source to that tag. Despite this workflow being manual, engineers already save significant time (up to 60%), when compared with alternative methods for creating S-P&amp;IDs.</p><p>CV, however, raises the bar even further.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IV97hLubooFI8cGj0572Qg.png" /><figcaption>Time savings realised by Clarity’s S-P&amp;ID generation for a set of P&amp;IDs, with and without Computer Vision automation</figcaption></figure><p><strong>This is how a CV-automated S-P&amp;ID generation process works:</strong></p><ol><li>A set of P&amp;IDs are user-uploaded to the platform for data extraction.</li><li>CV is used to extract component codes from the documents, after analysing a variety of text sizes, fonts, orientations, and arrangements (i.e. multiple lines, red-lined drawings, etc.).</li><li>Component codes are matched against component lists, with CV recognising links between name variations (e.g. “Pump-3” = “P3” = “P-3&quot; = “Pump 3”).</li><li>Component types are automatically classified using CV identification of pumps, motors, valves, and more.</li><li>Bounding boxes are digitally augmented onto the S-P&amp;IDs to highlight components.</li><li>Auto-tagged S-P&amp;IDs are finally reviewed by qualified engineers to ensure accuracy.</li></ol><blockquote><strong>CV reduces</strong> the time taken to create S-P&amp;IDs from <strong>12 hours</strong> per drawing, <strong>to</strong> less than <strong>30 minutes</strong> each.</blockquote><h3>Why S-P&amp;IDs really are a no-brainer</h3><p><strong>1Ease</strong> <strong>of</strong> <strong>use:</strong> P&amp;IDs typically come in two formats: oversimplified PDFs; or complex software-specific files. Neither is ideal for operational use. HyBird Clarity’s S-P&amp;IDs, however, are accessible via a simple online web-app, allowing you to view and modify them from anywhere.</p><p><strong>2 Do more faster:</strong> On average, engineers currently spend 60% of their time looking for data instead of using it. Conversely, an always updated single-source-of-truth can triple productivity. S-P&amp;IDs make finding equipment information and planning work straightforward.</p><p><strong>3Revision management:</strong> P&amp;IDs and other technical documents require regular updates. Instead of red-lining drawings or recreating them from scratch, an S-P&amp;ID simplifies revision management by allowing changes to be made and logged digitally.</p><h3>Here’s how you can make the switch</h3><p>It’s no secret that Smart P&amp;IDs make engineers’ and managers’ lives significantly easier, both on-site and in-the-office. Complex work plans and delayed decision-making are fast becoming a thing of the past.</p><p>If you’d like to find out how you too can implement S-P&amp;IDs, I think we might be able to help…</p><p>Drop me an email today to get started: <a href="/service/mailto:mohamed@hybirdtech.com">mohamed@hybirdtech.com</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=74f0aaa3cf8b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/hybird/why-your-organisation-should-ditch-p-ids-right-now-part-2-74f0aaa3cf8b">Why your organisation should ditch P&amp;IDs right now — Part 2</a> was originally published in <a href="https://medium.com/hybird">HyBird</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why your organisation should ditch P&IDs right now — Part 1]]></title>
            <link>https://medium.com/hybird/why-your-organisation-should-ditch-p-ids-right-now-part-1-d210d90e655b?source=rss----c70a352ce428---4</link>
            <guid isPermaLink="false">https://medium.com/p/d210d90e655b</guid>
            <category><![CDATA[oil-and-gas]]></category>
            <category><![CDATA[computer-vision]]></category>
            <category><![CDATA[iot]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[asset-management]]></category>
            <dc:creator><![CDATA[Mohamed Hafez]]></dc:creator>
            <pubDate>Mon, 01 Mar 2021 10:30:29 GMT</pubDate>
            <atom:updated>2021-03-01T10:30:29.399Z</atom:updated>
            <content:encoded><![CDATA[<h3>Why your organisation should ditch P&amp;IDs right now — Part 1</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hEVQP6uinaR-9Ld-Dw5mBA.png" /><figcaption>Smart P&amp;ID produced using AI and Computer Vision.</figcaption></figure><h3><strong>We all know what P&amp;IDs are…</strong></h3><p>Every process industry operator uses detailed Piping and Instrumentation Diagrams (P&amp;IDs) for a systems-level overview of how plant components interact — typically identifying process equipment, instrumentation, and control devices.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ha9aGrXOg2K2X8UXEm0QlA.png" /><figcaption>Pesky P&amp;ID.</figcaption></figure><p>These P&amp;IDs often reside on siloed information management systems (or even desks) in the form of 2D documents, such as PDFs, DWGs, or paper printouts.</p><h3>…and I bet you know why they’re doomed.</h3><p>This siloed, abstract data source that is the P&amp;ID is a crucial tool for managing complex plants, and yet it’s remained unchanged for decades. The information it provides is neither adequate nor accessible when it comes to critical decision-making for engineers and other site personnel.</p><h4>Disparate Data</h4><p>It’ll come as no surprise that trying to access a myriad of on-prem data systems (some of which are outdated) then relying on multiple site stakeholders to respond to queries leads to a huge waste of both time and money.</p><h4>Lack of Context</h4><p>On top of missing or inaccurate information, engineers must suffer the challenges of converting complex industrial facilities into simplified 2D drawings, only to then re-interpret the drawings so that site works can be planned and executed.</p><h4>Poor Knowledge Transfer</h4><p>Without accurate 3D location information, costly errors are easy to make — staff, contractors and other third parties aren’t able to tap into the tacit knowledge held by site specialists.</p><p><strong><em>Unfortunately, however, this is how many industrial facilities still operate today.</em></strong></p><h3><strong>So, why do we bother using them?</strong></h3><p>Well, the main functions of the P&amp;ID are:</p><ul><li>To <strong>identify</strong> <strong>components</strong> &amp; instruments</li><li>To <strong>show</strong> <strong>connections</strong> between components &amp; instruments</li><li>To <strong>visualise</strong> <strong>locations</strong> of components &amp; instrument</li><li>To <strong>illustrate</strong> the <strong>functions</strong> of components &amp; instruments within a process</li></ul><h3><strong>But what will we replace them with?</strong></h3><p>A system that provides all of the <strong>necessary site information</strong> needed by engineers and managers, in one single-source-of-truth:</p><p><strong>The Smart P&amp;ID (S-P&amp;ID).</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MrkZ6s1WcLHo5hWgX2CMAQ.png" /><figcaption>S-P&amp;IDs consolidate and contextualise site data, to make work easier.</figcaption></figure><blockquote>Smart P&amp;IDs allow engineers to access all necessary site information from <strong>one, always up-to-date, digital hub</strong>.</blockquote><p>By clicking on any component within an S-P&amp;ID, users can instantly access every bit of information about that particular component. This typically includes:</p><ul><li>Component details</li><li>Real-time gauge readings</li><li>Historical sensor readings</li><li>Predictive analytics insights</li><li>Operations and maintenance manuals</li><li>Inspection and maintenance reports</li><li>Work orders and plans</li><li>3D component location</li><li>Component CAD model</li></ul><p>These S-P&amp;IDs aren’t limited to the above; they can also be customised to integrate any additional data sources — helping engineers make more informed decisions <strong>quicker</strong> and <strong>easier</strong>.</p><p>At this point, you’re probably wondering how you can make the switch from conventional P&amp;IDs to Smart P&amp;IDs.</p><p>Well, look no further, we cover that right here in <a href="https://medium.com/p/74f0aaa3cf8b">part 2</a> of the blog.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d210d90e655b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/hybird/why-your-organisation-should-ditch-p-ids-right-now-part-1-d210d90e655b">Why your organisation should ditch P&amp;IDs right now — Part 1</a> was originally published in <a href="https://medium.com/hybird">HyBird</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What every oil & gas operator should know about Computer Vision and AI]]></title>
            <link>https://medium.com/hybird/what-every-oil-gas-operator-should-know-about-computer-vision-and-ai-45c8782eac31?source=rss----c70a352ce428---4</link>
            <guid isPermaLink="false">https://medium.com/p/45c8782eac31</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[asset-management]]></category>
            <category><![CDATA[computer-vision]]></category>
            <category><![CDATA[oil-and-gas]]></category>
            <category><![CDATA[iot]]></category>
            <dc:creator><![CDATA[Ahmed Hadid]]></dc:creator>
            <pubDate>Mon, 11 Jan 2021 13:19:16 GMT</pubDate>
            <atom:updated>2021-01-11T13:19:16.477Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PiLOcuBvm7Pj3D17TR473A.png" /><figcaption>Computer Vision-generated 3D Reality Model of an oil terminal (<strong>not a photo</strong>).</figcaption></figure><p><em>AI, Computer Vision, and Machine Vision</em> — I’m sure you’ve heard all these buzzwords before. Almost every “technology” and “innovation” department within the heavy industry sectors is on the hunt for AI solutions to automate operations and maintenance procedures that they barely understand, to achieve results that they’re not accountable for.</p><p>This begs the question:</p><blockquote><strong>Should your organisation even be looking at AI and Computer Vision at all?</strong></blockquote><p>The short answer is… <strong><em>Yes</em></strong>.</p><p>But before diving deeper, let’s clarify what they are:</p><ul><li><strong>Artificial Intelligence</strong> (AI) is a branch of computer science that focuses on building smart machines capable of performing tasks that typically require human intelligence.</li><li><strong>Computer Vision</strong> (also known as Machine Vision) is an interdisciplinary field of AI concerned with providing computers, machines, and robots with the ability to understand digital images and videos.</li></ul><p>AI and Computer Vision are powerful technologies that can help you improve your business in many ways. Here are the top 3 benefits driving its adoption throughout the oil and gas industry:</p><h3><strong><em>#1 — Grow your top- and bottom-line.</em></strong></h3><p>Over the past 12 months, almost every oil and gas operator has suffered devastating financial losses due to COVID-19 and the oil pricing crises. When implemented properly, AI and Computer Vision technologies can mitigate part of this financial risk. Novel analysis techniques can provide executives and managers with the ability to maximise their throughput, globally, whilst minimising the total cost of operations and maintenance.</p><p>Computer Vision can produce digital 3D Reality Models (3DRMs) of facilities to collaboratively visualise entire asset portfolios and intuitively understand detailed site information, from anywhere in the world.</p><p>When combined with additional datasets (such as equipment health, environmental information, and market activity), AI analysis can generate smart insights to streamline and improve board-level decision-making for corporate and commercial activities.</p><p>This solution can enable business continuity even during pandemics and downturns, whilst maximising both revenue and profitability.</p><h3><strong><em>#2 — Fulfil the commercial potential of your entire asset.</em></strong></h3><p>An asset’s commercial performance is measured by its overall throughput, which is a function of both asset availability and asset reliability. Its commercial potential is therefore realised when the asset is maintained optimally — and a truly optimised asset maintenance solution is cost-efficient and balances asset risk mitigation with financial performance. Reliability-Centred Maintenance (RCM) is the strategy that enables this.</p><p>RCM, however, is not achieved using just IoT sensors and equipment alerts.</p><p>The most effective solution, in fact, relies on AI-powered predictive equipment failure insights integrated into an advanced asset management platform, which visualises component-level 3DRMs, analyses intelligent P&amp;IDs, and automatically creates remedial work plans for site personnel to execute.</p><p>This Prescriptive Maintenance (RxM) approach is redefining asset maintenance and has proven to reduce unplanned downtime by up to 60%, increase asset availability by up to 10%, and reduce total maintenance costs by up to 30%.</p><h3><strong><em>#3 — Achieve a highest quartile safety record.</em></strong></h3><p>Workforce safety is a necessary focus for all oil and gas operators, given the myriad of site hazards and extreme environments that work must be completed within. Poor safety standards and inadequate maintenance procedures have continued to cost human lives for over a century now, and these incidents rightfully result in huge fines from HSE regulators. Recently, however, AI and Computer Vision algorithms are helping operators minimise accidents on site.</p><p>Video feeds from CCTV and retrofitted cameras are analysed by Computer Vision software to automatically to assess PPE compliance and record violations such as engineers not wearing helmets and other equipment correctly (or even at all). These insights can then be used to train site personnel to properly protect themselves, in case of any facility issues.</p><p>Instead of relying on periodic walkarounds for gauge measurement readings, Computer Vision also has the power to digitalise analogue gauges and meters, so a continuous live stream can be viewed and analysed intelligently without setting foot on site.</p><p>Not only is RCM the solution for maximising an asset’s commercial potential, but it also increases overall site safety, resulting in fewer unplanned equipment failures, and providing clear procedures on how to remedy any issues safely and efficiently.</p><p>These benefits aren’t always so trivial to achieve, though. In fact, here are the top 3 reasons that AI adoption could fail within your business:</p><h3><strong><em>#1 — Your organisation isn’t ready for new technology.</em></strong></h3><p>Despite the huge realisable benefits, most oil and gas organisations really aren’t ready for new technology — and it’s because of these two words: <strong>influence</strong> and <strong>accountability</strong>.</p><p>Successful technology deployment requires a <strong><em>champion</em></strong> from within the business, who:</p><ul><li>is a strategic and creative thinker, with an execution-oriented approach;</li><li>has enough influence to convince the board, executives, managers, and engineers that their suggestions should be followed;</li><li>is ultimately responsible for the successful deployment of the new solution; and</li><li>is accountable for the business’ P&amp;L (or other success metrics).</li></ul><p>Without the above requirements being satisfied, any new enterprise-wide innovation and change management is doomed to fail.</p><h3><strong><em>#2 — It can cost a fortune.</em></strong></h3><p>With all the hype around “AI”, it can be challenging to identify expert vendors that can really achieve the full set of Computer Vision and AI benefits for your business. Remember, though, buying AI technology is just like buying anything else: <strong><em>you get what you pay for</em></strong>. And if you obsess over price instead of value, there’s a good chance you can kiss your ROI goodbye.</p><p>Companies often look to save money upfront and end up selecting cowboy vendors instead of experts, then watch their solution fail, and repeat this same process two or three more times with the same result. This eventually results in overspending and achieving no ROI.</p><p>On the other hand, some asset operators blindly trust vendors, who abuse this trust and attempt to oversell them with every possible technology solution under the sun. While this might yield better results than the cowboy vendor, it does so at the expense of an optimal ROI, and with a <em>huge</em> upfront cost.</p><p>The best approach is to work with a vendor that is a technology and domain expert, with demonstrable experience in deployments for similar businesses and assets, and who treats you like a trusted partner.</p><p>If you buy right the first time, it will be an investment in your business’ profitability and longevity.</p><h3><strong><em>#3 — Your staff are afraid of change.</em></strong></h3><p>Today’s corporate inertia (resistance to change) is a sure-fire way of achieving tomorrow’s insolvency, particularly when operating in a dynamically priced market. And yet, I’m always astounded by how many heavy industry executives, managers, and engineers tell me:</p><blockquote><strong>“If it ain’t broke, don’t fix it.”</strong></blockquote><p>I’m confident that these famous words have guided corporate strategies for a copious number of (major) companies worldwide, and I’m even more confident that most no longer exist today. Those that do exist are now so small that we’ve simply forgotten about them. Think Blackberry, Blockbuster, Nokia, Kodak, AOL, and a whole heap of others (including thousands of oil and gas businesses that are wiped out with every downturn).</p><p>Future-proofing your operations demands continuous innovation, and innovation requires effective change management. So, why exactly is change so difficult?</p><p><strong><em>Executives and managers say:</em></strong></p><ul><li>They’re afraid of failure.</li><li>They’re afraid it will cost too much.</li></ul><p><strong><em>Engineers and site personnel say:</em></strong></p><ul><li>They’re afraid it will require too much effort.</li><li>They’re afraid of being laid off.</li></ul><p>After sharing successful case studies, most people are generally convinced. The hardest challenge, however, is that people are inherently lazy. It doesn’t matter how good or fancy your technology stack is. If nobody wants to use it, it will fail.</p><p>It’s important to remember that AI and Computer Vision technologies actually enable teams to work smarter, without working harder. They can consistently achieve better, more impactful results in a fraction of the time.</p><p>The upsides of AI and Computer Vision for oil and gas businesses are clear, however, the answer to the extended question of whether you should actually be using AI and Computer Vision is that <strong><em>it depends</em></strong>.</p><p>It depends on if your organisation and its people are <em>really</em> ready for change, and it depends on if you know who to work with to achieve that change efficiently and cost-effectively.</p><p>If you’d like to explore how AI and Computer Vision can work for your oil and gas business, we’d be happy to <a href="https://www.hybirdtech.com/contact/">arrange a call with one of HyBird’s experts</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=45c8782eac31" width="1" height="1" alt=""><hr><p><a href="https://medium.com/hybird/what-every-oil-gas-operator-should-know-about-computer-vision-and-ai-45c8782eac31">What every oil &amp; gas operator should know about Computer Vision and AI</a> was originally published in <a href="https://medium.com/hybird">HyBird</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>