By Michael Ford, European Marketing Director, Aegis Software Corporation
Numerous in-house software systems, which includes externally developed bespoke systems, have become a part of the critical manufacturing support infrastructure in most assembly factories. These systems are essential components of engineering, quality, and the manufacturing flows. As factories now face a step-change challenge to move forward with Smart digital manufacturing that Industry 4.0 represents, can existing in-house solutions be adapted, or are digital factory best practices achieved with the introduction of a new commercially developed digital MES solution?
Changing Industry Requirement:
In-house “point solution” software applications were created as a result of specific needs within manufacturing, often tied to specific processes or hardware, and are based on the software technology of that time. Many generations of point solutions continue to work alongside each other in most factories, after all, if it isn’t broken, don’t fix it. The production models and practices behind point solutions however, as well as legacy MES (Manufacturing Execution Systems), ostensibly originated around high-volume mass production. An isolated set of engineering data for a product that would run steady-state for long periods with a predefined configuration and the push-delivery of material kits and resource assignments satisfied requirements. A steadily increasing demand for the factory to produce a higher mix of products, together with smaller lot sizes, has put progressively increasing strain on these legacy solutions, resulting in the need in many cases for urgent “quick and dirty” modifications. Being able to cope with changing requirements in production via everincreasing customizations by the in-house IT department is however a long way from providing flexibility in line with the latest digital best practices. As the industry moves forward toward the ultimate model of flexibility & automatic adaptability that Industry 4.0 represents, manufacturing is expected to produce each day only what the customers want for immediate delivery, without loss of productivity. The goals today are elegant and automatic single-piece flow and ultimate configureto- order. Legacy software solutions cannot easily cope with this scenario, creating the need for a completely new technology and architecture. The situation is rather like replacing one’s CD and DVD collection with digital media streaming services; there are fundamental differences in supporting requirements.
A steadily increasing demand for the factory to produce a higher mix of products, together with smaller lot sizes, has put progressively increasing strain on these legacy solutions
The Digital Streaming Factory:
An Industry 4.0 factory has many things in common with a media streaming service. Customers are expecting to get what they want, in any quantity they want, whenever they want, whilst reserving the right to frequently change their minds and ask for something different. The Smart assembly factory has to be able to respond immediately in these situations, without the overhead of a warehouse full of finished goods making it appear that the factory can provide flexible delivery. Immediate response in the factory operation requires immediate decision-making across multiple disciplines. Whereas traditional factory practice is to simulate, plan and optimize scheduled operations weeks or even months in advance, plans now need to be created “on the fly”, repeatedly every day, and yet, also need to be optimized such that the impact of changeover losses are minimized. The human process of decision-making, based on discussion, meetings, phone calls, emails, and a quick Excel report, is simply too slow to support the Industry 4.0 factory, never mind provide the required optimization. Instead, relevant data about the progress and status of every process and operation must be available continuously, gathered and processed by intelligent software modules working together across the factory. The result is the creation of immediate decision-making support for managers and engineers, and, in an increasing number of cases, automated decision-making. There are two critical technological elements to the solution, needed irrespective of how the solution will be created, the use of IIoT technology and a digital infrastructure platform.
The Reality of IIoT Technology:
Transfer of data using IIoT technology allows information from any data provider to be available for use at any time, by any data consumer. Any provider may also be a consumer, as IIoT is bi-directional in nature. Unlike the definition of legacy bespoke interfaces created for specific uses, the definition of the data to be exchanged through the IIoT interface is based on the scope of the device rather than a specific application. Many different applications will be expected to use the IIoT data in parallel. As IIoT data will be exchanged between devices from many different vendors, it is critical that there is conformance to a universally agreed data definition standard. The three ingredients needed for an effective IIoT standard are the protocol, data encoding, and language definition. Take for example a cell-phone, where handsets from different vendors communicate in a standard way over a standard network. Telephone calls are however only effective where both people understand a common language. For IIoT systems, the issue is exactly the same. To date, legacy interface standards for data communication in assembly manufacturing, have neglected to fully address this final requirement, and have not been successful.
By contrast, the new IPC CFX (Connected Factory Exchange) IIoT standard consists of a clearly defined protocol (AMQP v1.0), data encoding method (JSON), as well as the comprehensive definition of standard content. CFX has been created through a collaborative consensus across the industry, as machine vendors, manufacturers and software solution providers anticipate being a part of the next generation of digital factories. The scope of CFX covers the entire range of assembly factory operations, including all forms of production assembly processes, test, inspection, supply chain, quality control, environment and planning. The specific focus of CFX is on the enablement of Industry 4.0 solutions, including provision for Augmented Reality (AR) and Artificial Intelligence (AI) technologies which are expected to be common within the industry over the coming years. Any Smart factory solution being considered needs to have the capacity and infrastructure to embrace the use of standard IIoT technology such as CFX.
CFX has been created through a collaborative consensus across the industry, as machine vendors, manufacturers and software solution providers anticipate being a part of the next generation of digital factories
The challenge for both in-house and the MES solutions going forward is the ability to work in the real-time IIoT messaging environment, rather than the traditional approach of gathering data from specific interfaces, some form of interpretation of content, the population of a series of databases, which then need to be combined somehow as reporting software attempts to piece data together to create actionable information. For most legacy systems, the use of IIoT data represents a critical paradigm shift in the way that they are architected, meaning that significant refactoring is required.
There is good reason however for the transition from legacy manufacturing solution technology, to IIoT based technologies. Rather than a narrowly focused needs-based design, IIoT provides the framework for any conceivable Industry 4.0 solution idea that comes along. IIoT in itself is the enabler of a rich and diverse range of solutions, as opposed to being a solution in itself. The degree of capability and value derived from IIoT depends on how the technology is applied, especially where multiple solutions from a user-perspective, will work together seamlessly on a single digital infrastructure.
Digital Infrastructure Platform:
One of the strengths of MES is the underlying infrastructure that models the manufacturing operation and provides the underlying structure on which MES functions work. This becomes very much more critical in an Industry 4.0 environment, where changes need to be executed urgently whilst bearing in mind any constraints, consequences and loss of production optimization. Product assignments to line configurations should be reviewed, including process engineering data creation, as well as the availability of materials, tools and other key resources. Decisions in the Industry 4.0 environment need to be taken quickly. The whole manufacturing operation needs to seamlessly transition from a current plan to a new plan. The common digital infrastructure that inter-connects processes and functions is absolutely mandatory. The scope of decision consideration may also extend to interact with machine vendors’ Industry 4.0 environments, which also adapt and optimize accordingly.
The connection and integration of all of the elements within manufacturing is a fundamental difference from the way in which the majority of legacy point solutions have been developed. Even legacy MES providers are significantly challenged, as many are made up of different historical components linked together, with bolt-on applications that are not fully integrated. This scenario may have been a reasonable compromise for legacy production practices, but fails to provide support for the latest and future generations of digital manufacturing with Industry 4.0.
Decisions in the Industry 4.0 environment need to be taken quickly. The whole manufacturing operation needs to seamlessly transition from a current plan to a new plan
Build vs Buy:
For an in-house software development team, the realization of the infrastructure required relating to workorders, shift patterns, machine and line configurations, engineering data, work assignment, tools, maintenance regimes, quality control, conformance, compliance and traceability, together with the introduction of IIoT technology, will in all cases mean a complete paradigm change. Faced with outdated software systems, the most common decision will be to recycle the accumulated manufacturing knowledge, but none of the code. A clean sweep and unencumbered start will be an attractive option, utilizing the latest software technologies and standards such as CFX. The in-house team will get the opportunity of seeing and understanding the exact new requirements and needs of manufacturing. Getting up to speed however to build a risk-free solution internally, with the required infrastructure to provide advanced, realtime decision-making functionality, will take a great deal of time and investment. The delay until return on that investment is achieved may run into several years. Maintaining professional grade internal resources for such projects is a major expense for a manufacturing company, which adversely affects business performance metrics.
Within the MES industry, unfortunately, the majority of solutions will face similar technical issues as the in-house system developers. Whether the legacy MES infrastructure of each solution can be adapted to work in the digital IIoT environment will vary from case to case, but many compromises can be expected as a result. The effectiveness of systems in such cases will be difficult for customers to immediately understand until the solutions have been deployed and in use for some time. Many functions appear to look similar across solutions but in fact are often completely different in the way that they work and the values they create. In terms of business risk, having a solution on-hand, having been developed and tested professionally, in use at other customers, and being immediately available for installation however, represents a lower business risk, a shorter time to value and a quicker return on investment, with little effect on the fixed costs of an operation as compared with a dedicated software team.