Using contractors involves an outside organization that will create risk to the site company. The contractors who are unfamiliar with the facility may create process hazards to the company. Thus companies must recognize and address challenges associated with using contractors; and select contractors based on stringent criteria. Only then can the company ensure safety of all people (onsite and offsite).
Contractor management is a system of controls to ensure that contracted services support safe facility operations. This element addresses the selection, acquisition, use, and monitoring of such contracted services. Systems must be established for qualifying firms based upon not only their technical capabilities, but also their safety programs and safety records.
The boundaries of authority and responsibilities must be clearly defined for any contractor that works at the facility. Periodic monitoring of contractor safetyperformance and auditing of contractor management systems is required. After completion of the work, evaluation of the safetyperformance should help to determine whether the contractor can be used again in any future work.
Contractors are not familiar with the facility safety controls and procedures hence the company needs to train them to understand the safety controls and procedures prior to them starting their work. The company must also make contractor to aware that they treat safety seriously and the contractors must adhere to all safety rules, guidelines, procedures, etc. while performing their job. Training must be provided to contractors if they are handling critical tasks that may create major process safety hazards.
Don’t treat contractor management as a petty issue, it can create process safety hazards if it is not well managed.
Artificial intelligence (AI) and machine learning already impact our everyday lives. These technologies mostly work so seamlessly with our daily experiences that we barely notice. For instance, machine intelligence powers the digital assistant people use on their phones, movie suggestions on streaming websites, and filters in email. In another generation, it’s possible to imagine that people will consider such revolutions as self-driving cars just as ordinary as their recommended movies on Netflix. Yet, according to CIO Magazine, AI and machine learning are just now making inroads into corporate IT departments.
AI, Machine Learning, and the Changing Roles of Chief Information Officers
In the CIO article cited above, Dave Wright serves as the CIO of Service Now. He said, in the past, his role as Chief Information Officer served to guide, build, and maintain a company’s tech infrastructure. These days, that role has evolved to focus more upon strategizing ways to use technology to benefit his organization. For instance, the CIO may not always choose to expand or even keep their own internal IT infrastructure as much as survey existing technology to see what the business can use to meet its business goals.
Sometimes, this role may involve shrinking the company’s own computing power and partnering more with providers who offer the best solutions. If companies don’t have the resources to develop their own intelligent systems, they can rely upon trusted third parties for solutions.
Facing Internal Resistance to Artificial Intelligence
Wright understands that some members of the IT or other departments may fear changes, specifically the adoption of machine learning and AI. They have concerns that they will detract from their own duties. As was the case during the first days of digital transformation”>digital transformation, these technologies seldom remove jobs but allow people performing those functions to work more productively in a way that supports their organization’s true business goals. While it’s up to CIOs to explore solutions, they also need to communicate the benefits of those solutions to their employees and other executives.
AI’s Penetration into Today’s Businesses
Right now, according to the survey CIO Magazine reported upon, almost 90% of companies do use AI and machine learning in some fashion. However, about 66% of the businesses surveyed are only researching or piloting these new smart technologies. Only about 23% responded that they either used machine intelligence either in several parts of all of their business. Wright believes that most businesses will start by using AI to help interpret and organize information. Only after they feel comfortable with that aspect, will more businesses move to using it to solve problems and later, to anticipate and remediate them.
AI vs. Machine Intelligence
As a note, sometimes people use artificial intelligence and machine learning interchangeably. Artificial intelligence describes computer systems that use their algorithms to mimic human decision-making ability. Machine learning describes a type of artificial intelligence that can use information to adapt its algorithm based upon the information that it receives. In that way, machine learning refers to a kind of artificial intelligence.
Why Cleaning Up Bad Data Matters for Effectively Using AI
As we experience the Information Age, people sometimes refer to data as the new “oil” because of its value. As companies collect more and more information, they increasingly wrestle with problems of data quality. Without proper management, information gets corrupted because its obsolete, redundant, or simply in error. As CMSWire pointed out, new compliance rules like GDPR and the California Consumer Privacy Act also can turn the problem into a regulatory hazard.
Mostly, Wright emphasized that artificial intelligence and machine learning technologies will only work as well as the information that they receive. Luckily, businesses can find an intelligent solution to help them with that task as well. Gartner says that machine-augmented data management will grow common within many organizations. In other words, AI can provide the solution to making certain that it can get the best possible information to base its processing upon.
Why Are CIOs Exploring the Benefits of AI and Machine Intelligence?
Increasingly, high-tech companies are offering smart features to consumers to help them make better and faster choices or to do things more efficiently and safely. It only makes sense that businesses can find plenty of ways to employ this technology to improve their own business processes. Even better, AI tech providers can help level the playing field, so that businesses without the resources to develop their own tech can still access it. These intelligent machines can help companies reduce threats and enjoy more value from the increasingly large amounts of information that they collect.
Technology often gets all the credit when it comes to predictive maintenance. That’s unfair. Technology is important, but it’s only a small piece of a great predictive maintenance program.
In short, the journey to predictive maintenance is slow, but worth it if it’s done right.
This article is all about building a predictive maintenance program that will last. We explore the six pillars of a strong predictive maintenance program, how you can develop each area, and how to use them to achieve predictive maintenance.
A short refresher on predictive maintenance
Predictive maintenance (PdM) lives in the same family as preventive maintenance. They’re both proactive types of maintenance—work is done on an asset before something bad happens to it, not after a failure has shut it down.
The difference between preventive maintenance and predictive maintenance lies in the methods used, the amount of lead-time you have for a task, and the precision of scheduling. PdM uses condition-monitoring tools and techniques and asset information to track the real-time and historical equipment performance so you can anticipate failure before it happens.
Since predictive maintenance aims to give you an ideal window for proactive maintenance tasks, it can help minimize the time equipment is being maintained, the production hours lost to maintenance, and the cost of spare parts and supplies. We outline maintenance-strategies/”>where predictive maintenance fits in your overall maintenance strategy here.
The six pillars of a predictive maintenance program
A sturdy predictive maintenance program is built on six pillars: People, data, processes, tools and parts, equipment, and technology. If one pillar is not stable or is left to rot, your whole program can crumble.
People: Culture eats strategy for breakfast
The long journey to predictive maintenance always starts with people.
“It doesn’t matter if your predictive maintenance plan looks good on paper if you don’t have buy-in from the people who are doing the work,” says Fiix’s solutions engineer Jason Afara.
“In other words, culture eats strategy for breakfast.”
Every other pillar of a predictive maintenance program needs people to build and maintain it. Data needs interpreting. Technology needs setting up and managing. That’s why everyone in your organization should understand how PdM works, why it’s important and what they can do to make it successful.
Getting people at your facility onboard with the (many) changes that come with predictive maintenance is absolutely essential, but not always easy. This article from Software Advice offers some great tips on change management, getting buy-in from your maintenance team, and creating a great culture at your facility.
Technology is like a dash of salt in a predictive maintenance program—it ties the other ingredients together and makes them shine.
“Without the data, you can’t predict anything. If you don’t have a baseline about what’s normal for a pump or a conveyor, you can’t identify or predict anomalies,” says Bryan Sapot, CEO of SensrTrx.
But with quantity also comes the need for quality.
“If you don’t have good information coming from the plant floor, it won’t matter how good your algorithms are, you won’t be able to make good decisions with it,” says Jared Evans, the chief operating officer at MAJiK Systems.
Data is the link between current assetperformance and the future state of the asset. That’s why everything, from throughput to failure modes and beyond, must be constantly updated. These numbers also have to be accurate everywhere. If they’re different from system to system, it’ll throw your whole program into disarray.
Or as Jason puts it, “If you have bad data coming from your machines and software, it’s like the weatherperson telling you it’s sunny out when it’s actually raining. You’ll step into the rain and get soaked.”
Processes: A steady hand on the predictive maintenance ship
Simply put, your processes are the way you work your plans and does the things it needs to do every day to be successful. An effective predictive maintenance program helps make your whole operation predictable so it can maximize everything from working hours to assetperformance.
Processes in a predictive maintenance program are people-driven and equipment-driven.
People processes involve the way your maintenance team goes about their work. They outline how staff interact with machines, data, each other, and everything else.
“You need to understand who is responsible for what, how frequently you review data and tasks, how you communicate, and how you plan, escalate, and complete tasks,” says Jason.
When it comes to equipment processes, Jared says it’s crucial to know what processes your equipment completes, how to capture assetdata, and how the data maps to future performance.
If you have bad data coming from your machines and software, it’s like the weatherperson telling you it’s sunny out when it’s actually raining. You’ll step into the rain and get soaked.
Tools and parts: Trusty sidekicks ready for the spotlight
Tools and parts play a huge role in allowing predictive maintenance to go from a far-away dream to a realistic goal.
“Predictive maintenance isn’t new,” says Jason. “The difference between 20 or 30 years ago and now is that we have the tools and understanding of parts to do it better and with fewer costs.”
Tools are the instruments used to measure the condition of assets, like infrared cameras, and the tools needed to inspect or repair equipment. Parts are the different components of equipment, but not just any old parts will do for predictive maintenance, as we’ll see below.
Equipment: Not all machines were made for predictive maintenance
Anyone who says reactive maintenance can be totally eliminated has never had their windshield cracked by a stray pebble. While this isn’t exactly an on-the-job example, the lesson still applies to the shop floor: You can’t anticipate everything.
It’s important to know which of your equipment allows you to anticipate failure on it when setting up a predictive maintenance program
“The assets that fit into a predictive maintenance program are the ones that provide good condition data with enough lead-time to catch problems before total failure,” says Jason.
Jason also recommends applying predictive maintenance to your most critical assets with the most observable failure modes because of the time and money needed to build a PdM program.
If you’re looking for information on choosing the best equipment for your predictive maintenance program, check out this starter pack of resources (PF track with balanced maintenance strategies, P-F curve, condition-based maintenance blogs)
Technology: The glue that keeps the other elements together
Technology is like a dash of salt in a predictive maintenance program—it ties the other ingredients together and makes them shine. It helps you manage, facilitate, and optimize the other pillars of predictive maintenance.
“Technology gives you an extra set of eyes,” says Bryan, “so you can collect real-time data without having someone on your team constantly looking at the information.”
This is a big job, one that can’t be done by a single piece of technology.
“Predictive maintenance requires you to pull together so many different data sources,” says Jared.
“You need to know what products are being run and when, the cost of all your activities, when maintenance was last done. The list goes on. You need several pieces of technology to capture all this data, store it, and make sense of it.”
There are lots of different technologies that can be used to manage a predictive maintenance program, from ERPs to MES systems and CMMS software. We explored the most common of these technologies here.
How to build a predictive maintenance program
Predictive maintenance: Part of a balanced strategy
The best way to think of predictive maintenance is like a bowl of cereal in an old TV commercial: It’s part of a balanced breakfast (or maintenance strategy). Predictive maintenance isn’t the only strategy to strive for. Instead, it should supplement your overall maintenance program.
“Predictive maintenance will never replace all other forms of maintenance,” says Jason.
“Creating a predictive maintenance program isn’t about making a checklist. You can’t just tick off a bunch of tasks, flip a few switches and be completely predictive. It’s a journey. It might take 10 years to go 10% predictive.”
A predictive maintenance program won’t solve all your problems. But there are some serious benefits to having one, like a more reliable operation that allows everyone at your organization to grow and be more efficient.
Taking advantage of those benefits relies on building on key maintenance fundamentals. When those fundamentals are strong, you’ll have a strategy that’ll weather any challenge thrown at it.
What are the requirements of a Mechanical Integrity program?
Mechanical Integrity (MI) is significant in terms of the asset coverage involved. System examples include fixed equipment such as pressure vessels and storage tanks, piping systems and associated hardware (valves, fittings, etc.), relief devices, vent hardware, emergency shutdown/control systems. In many cases, this means that all equipment within the boundaries of a facility is subject to the PSM – Mechanical Integrity standard.
MI encompasses the activities necessary to ensure that equipment/assets are designed, fabricated, installed, operated and maintained to a desired performance in a safe, environmentally protected and reliable fashion. MI is a sub-set of an effective reliability program and overall asset management.
Let’s take a look at the MI requirements: –
Mechanical integrity (MI) of equipment has been controlled at all industrial facilities for many decades. Everything wears out eventually. With MI, it can help determine when that “eventually” might be (the operational life expectancy).
How to use Process Safety Analysis to the benefit of your process plant?
What is Process Hazard Analysis (PHA)? A PHA is required for any industrial process that makes use of hazardous chemicals. Its purpose is to identify the significance of scenarios (potential causes and consequences) that could result in fires, explosions, chemical spills and the release of toxic chemicals. It focuses on factors that might affect the process (equipment, instrumentation, utilities, human actions (routine and non-routine), and external factors).
Steps in the PHA Process
Methods for Conducting the PHA:
What-if Study – for review of an uncomplicated processes;
Checklist – for a more complicated process using a checklist;
Hazard and Operability Study (HAZOP) – a structured method to analyze possible deviations in design conditions;
Failure Mode and Effects Analysis (FMEA) – a systematic study of component failures that could conceivably affect the safety of the operation;
Fault Tree Analysis – either a qualitative or a quantitative model of all the undesirable outcomes, that could result from a specific initiating event; or
An appropriate equivalent method.
The process hazard analysis is best performed by a team with expertise in engineering and process operations. The PHA team should include:
Employees who have experience with and knowledge of the process being evaluated; and
Team leader who has knowledgeable in the specific PSM analysis methodology being used in the evaluation.
It is advised that, at least every five years after the completion of the initial process hazard analysis or whenever there is a change in process, the process hazard analysis must be updated and revalidated by a qualified team to ensure that the hazard analysis is consistent with the current process.
Recent Comments