Some of the concepts we will discuss below may sound a bit futuristic, bringing to your plant floor scenes from popular science fiction movies.
You may wonder if smart machines could someday exhibit scary abilities that would enable them to outthink and manipulate humans, which underlines the importance of the security and trustworthiness of smart machine deployments.
You certainly do not want your maintenance deployment to end up like the replicants of the popular Blade Runner film, which deviated from their original mission.
Nevertheless, as presented in previous posts, most of the technology pieces are already part of the ongoing digitization of industry. It’s really time to consider what features of smart machines are you using today, but also how to best leverage the benefits of smart machines and smart equipment as the latter become available.
Future maintenance activities will rely on a wave of IT innovations in the areas of BigData, Machine Learning, AI and the industrial Internet of Things (IoT).
These technologies will enable increasingly flexible and more intelligent maintenance operations. Along with novel IT technologies, future plants will also deploy smart machines and smart equipment, such as smart objects with embedded intelligence.
Smart machines will be able to host part of the application logic of the maintenance process, through predicting their lifetime and scheduling maintenance tasks. In this way, they will also increase automation and maximize the effectiveness of maintenance processes.
In particular, smart machines and smart equipment enable the next generation of IoT-based maintenance applications. This is accomplished through supporting not only applications involving passive sensors and centralized maintenance logic over their data, as well as more dynamic maintenance systems that can distribute part of their intelligence in the machines and the equipment as well.
As part of these maintenance applications, the role of humans will shift from manual tasks to more intelligent supervisory and/or decision-making tasks.
The characteristics of these Smart Machines
In principle, smart machines and smart equipment are able to act in an autonomous or semi-autonomous fashion by performing quite sophisticated tasks. Their main characteristics are as follows:
Smart machines are Cyber-Physical Systems (CPS): Smart machines will employ sensors and actuators in order to bridge the physical world (i.e. the plant floor) with IT processes. In this direction, they will also provide digital interfaces for interacting with humans (employees) and applications. An example of a CPS system is an intelligent workstation in a factory, which automatically adjusts to the application context and the ergonomics needed by its operator.
Smart machines and smart equipment will be intelligent: Unlike conventional machines, smart machines will incorporate advanced capabilities for process control and reasoning over the context of the target application, including reasoning towards optimizing maintenance tasks. These capabilities will be based on the ability of machines to process large amounts of information (typically BigData) quickly. In this context, they will also incorporate the latest advances in AI, which will enable them to reason about situations like humans do.
Smart machines will be interconnected: Contrary to past applications, machines and equipment will be interconnected, which will allow them to interact and exchange information about their status. This will enhance the intelligence of the machines, as illustrated in a TED talk by Markus Lorenz. The talk also explains the potential of interconnecting smart machines such as industrial robots, manufacturing machines, smart sensors and more.
Smart machines and equipment will host part of the maintenance logic: The logic of maintenance applications will no longer be implemented in a central point. Instead, machines and equipment will be capable of autonomously processing maintenance-related information and make relevant decisions based on information about the status of the plant, including other machines, equipment and their components.
Smart machines will change the roles of human workers in the maintenance process: Despite the rise of intelligent CPS systems, we still think of smart machines as being limited in providing recommendations about optimal course of action, while expecting that a human employee will make the final decision. The rise of smart machines will gradually change this paradigm through a shift toward machines that are able to understand the context of the maintenance task in order to engage in related transactions, such as scheduling maintenance activities and ordering spare parts.
Maintenance functionalities supported by smart machines
In the near future, the functionalities and characteristics of smart machines and smart equipment will provide a compelling value proposition for industrial maintenance tasks, through facilitating the following processes:
Automatic Data Collection: Data collection from sensors is nowadays one of the big headaches for maintenance engineers, as most machines and plant floor equipment are not capable of providing maintenance datasets. To alleviate this problem, maintenance solution providers are implementing edge devices, which provide a low-level interface to field equipment in order to facilitate the data collection process. In essence, such edge devices are converting legacy machines and equipment to cyber-physical systems, while enabling information filtering in order to get rid of data that are not useful for the maintenance tasks. Smart machines will ease the process of data collection by removing the need to deploy edge devices. Machines will provide interfaces for direct access to these data, while at the same time processing this data within the machine, based on its embedded intelligence capabilities.
Optimizing Operations through Self-Learning: Smart machines and equipment will be able to automatically learn how to optimize their operation through automatically correlating information about their operation, failures and end-of-life, with their performance and the quality of their operations (e.g., the quality of the products that they produce). Overall, machines will be able to learn how to operate in order to maximize their lifetime and the OEE (Overall Equipment Efficiency).
Predictive Maintenance: Predictive maintenance will be one of the core maintenance-related features of smart machines, which will be able to predict their lifetime. We have already explained the principles of predictive maintenance operations, where we also illustrated how parameters like EoL (End-of-Life) and MTBF (Mean-Time-Between-Failures) are calculated and used in the scope of FMEA (Failure Mode Effect Analysis) processes. Smart machines will comprise the intelligence needed to foretell their lifetime. They will also be able to assess different what-if maintenance scenarios in terms of their potential to maximize OEE.
Maintenance Scheduling and Optimization: The evaluation of different maintenance scenarios against their OEE can provide a basis for optimal scheduling of maintenance tasks. This means that smart machines and smart equipment will be able to identify the best time to schedule maintenance tasks, taking into account the workload of the plant (e.g., pending production orders or energy generation cycles). To this end, machines will be able to access required information from business information systems such as ERP (Enterprise Resource Planning), EMS (Energy Management Systems), Asset Management systems and more. Following the identification of optimal manufacturing schedules, machines will be also able to initiate maintenance related transactions such as ordering spare parts in ERPs and scheduling tasks for technicians in project management systems.
Smart Contracts: Recent advances in peer-to-peer systems for distributed transactions provide the means of executing maintenance related transactions in line with smart contracts (e.g., Service Level Agreements) between the parties. This will enable smart machines to order spare parts and plan maintenance tasks in-line with Service Level Agreements (SLAs). The technology that enables enforcement of smart contracts is the blockchain, which underpins the popular Bitcoin cryptocurrency. In future posts, we will discuss blockchain technologies and how they could be used for maintenance tasks.
A typical maintenance scenario involving smart equipment could combine all of the above functionalities. Hence, a machine will be able to collect data about its state, process the data in order to predict its end-of-life and to decide the maintenance schedule that maximizes OEE. At the same time, the machine will be able to initiate orders of spare parts in line with equipment vendor SLAs.
In this scenario, humans will no longer need to engage in manual data entry and contract enforcement tasks, which can be error prone. Nevertheless, humans are likely to have new roles in the process of building, deploying and supervising the operation of smart machines.
The benefits and the human factor
The rise of smart and connected equipment will deliver a host of benefits for maintenance stakeholders including:
Optimized OEE and reduced equipment costs, through facilitating timely repairs that reduce costs for components, while at the same time rendering the repair processes less laborious.
Optimized labor costs, due to the need of replacing specific components of the smart machine instead of the entire equipment, but also due to a smaller frequency of repairs for critical equipment (i.e. reduced “critical callouts”).
Increased productivity for employees, given that machines will instruct them to replace the right component at the right time, while also disrupting operations for a shorter time than in the past. Hence, planned downtimes will be reduced, while employees’ time and efforts will be better utilized.
Increased safety, as addressing potential problems earlier will ensure that employees work under safer conditions.
Most of the above benefits enable us to work safely and be more productive.
Author: John Soldatos
John Soldatos holds a Phd in Electrical & Computer Engineering. He is co-founder of the open source platform OpenIoT and has had a leading role in over 15 Internet-of-Things & BigData projects in manufacturing, logistics, smart energy, smart cities and healthcare. He has published more than 150 articles in international journals, books and conference proceedings, while he has authored numerous technical articles and blogs posts in the areas of IoT, cloud computing and BigData. He has recently edited and co-authored the book “Building Blocks for IoT Analytics”.