GUIDE TO Model-based Working Process within LKAB
1
Photo: Fredric Alm, Alm & ME
Preface LKAB’s future depends on the affordances of technology and process development, as the company is facing major challenges. Ore must be mined at even greater depths and at lower costs. New and more complex minerals are to be processed, energy usage must be more efficient and fossil fuels are to be phased out. Environmental and climate impacts must be reduced. Logistics and transport to ports must be made more competitive. Work methods should be characterized by efficiency, commitment and innovation. Increased digitization permeates all processes and is necessary to achieve high efficiency and competitiveness. A current digitization initiative within LKAB has been established to encourage the use of digital models for more efficient management of increasingly complex operations. This guide describes whether and when a model-based engineering approach should be used and how the methodology can be implemented and further developed.
2
Terminology and Definitions WHAT IS A MODEL? • A simplified version of a concept, phenomenon, relationship, structure or system • A graphic, mathematical or physical representation • An abstraction of reality by eliminating redundant components • A model includes the following objectives:
- Facilitating comprehension - Supporting decision making - Enabling scenario analysis - Assuring predictability and controllability
WHAT IS A SYSTEM? A system is an arrangement of elements or parts where the holistic modus operandi surpasses those of the individual constituents by the added effects of positive synergies and interprocess benefits. (source: www.incose.org)
This guide was produced within an LKAB study named Model Based Engineering, April 2021. Authors: Göran Tuomas (goran.tuomas@lkab.com) and Peter Lingman (peter.lingman@optimation.se) Workgroup: Göran Tuomas (LKAB), Peter Lingman (Optimation), Kjell-Ove Mickelsson (LKAB), Håkan Tyni (LKAB), and Bo Lahti (LKAB).
3
Document-based vs. Model-based Engineering In principle, there are two fundamentally different ways of working with systems and information: model-based engineering (MBE 1 ) and document-based engineering (DBE 2 ). Both methods have their advantages and disadvantages. DBE is well known and established within LKAB but is best suited for smaller systems and organizations. MBE is suited for handling complexity and enables a more comprehensive systems analysis. This is critical for meeting the everchanging challenges in a changing world that rapidly moves toward intense digitalization. New skills, new organization and new mindsets are necessary in facing this endeavor. The application of traditional document-based engineering for the development and maintenance of complex systems can result in tens of thousands of documents being generated during a system’s lifecycle. These documents must be stored, structured, maintained and communicated to the right people for a seamless operation. Changes to one document may require manual work to update subsequent documentation. This process is not only tedious but is also susceptible to errors and faults. Furthermore, inadequate documentation may eventually lead to serious damage, failure, or extensive extra work to recreate information. Model-based engineering allows for the harmonization of a broad scope of information from different fields. Changes in data or models are automatically propagated to all relevant systems and their users; therefore, information will be consistent and updated. Models will also identify any shortcomings, enabling traceability and analysis of the modeled system.
Goals with Model-based Engineering
1 MBE’s Swedish terminology within LKAB is modellbaserat arbetssätt 2 DBE’s Swedish terminology within LKAB is dokumentbaserat arbetssätt
4
DOCUMENT-BASED ENGINEERING (DBE)
Within LKAB, DBE consolidates information from various individual types of documents. Descriptions of complex relationships, requirements and systems mean that the document contents also become complicated and extensive. However, a document-based approach does not mean that all information needs to be conveyed in text form. The documents can be supplemented with models that describe examples of design or more abstract phenomena. The focus is not on models and their holistic contributions to efficient operations within all of LKAB.
DBE – Implicit connections between documents/models
MODEL-BASED ENGINEERING (MBE )
Can LKAB be described as having a systematic, efficient, and common use of models throughout the company? This use begins in early concept phases and continues through all lifecycles of the system. Models are used to describe system requirements and design, disseminate information, analyze behavior, control processes, and verify and validate information. MBE methodology has much in common with what the literature calls model-based systems engineering (MBSE). However, MBE aims at a suitable balanced use of digital models and data for LKAB, rather than a formal method.
MBE – Organization uses conglomerative system models/data
5
MBE PRINCIPLES • Common data and models are developed, used, managed, and reused over time and throughout the lifecycle of the facility. • Common data and models are original. • Model-based tools are used in all relevant business areas and together create a unity that enables a complete system analysis. • Duplicate models are avoided. Different activities and platforms use the same models whenever possible. • Structured data sources enable the use of new technologies and methods. • Standardized data formats
enable exchange between different applications and platforms.
• Secure communication
between users, platforms, and systems.
6
General execution of data storage of model-based engineering (MBE)
Data is generated, reworked, and continuously saved during the development of a facility. Models are the most refined type of data and can be defined as descriptive , which describe a current situation, or predictive , with the ability to simulate behavior over time. When developing new systems, predictive models are used to simulate the systems’ technical function, production capacity, economic impact, etc. The use of models when a plant is in the production phase of its lifecycle uses their prediction ability to support reasoning, decisions and optimization. Models make it possible to reach conclusions based on historical data with results from, for example, scenario analyses with predictive models.
7
Choice of Working Method BACKGRUND – LKAB’s NEW PRODUCTION SYSTEM LKAB has developed its new strategy for the future and establishes three main goals for the transition: (1) set a new world standard for mining, (2) develop direct reduced iron produced with hydrogen, a climate-friendly option, and (3) produce vital materials via the extraction of critical minerals from mining waste. The restructuring will require investments of approximately SEK 10 - 20 billion per year for a period of approximately 15 to 20 years. All primary production systems must be reviewed and further developed to make this possible. In terms of the production process, the design for mining, logistics and infrastructure must be developed so that the technology and methods become significantly more efficient than today’s mining processes. The ore will be mined at even greater depths with the subsequent challenges in efficiency, safety, volume, and costs. In refinement , new efficient refinement processes will be developed to meet future requirements for fossil-free direct reduced iron production. Increasingly stringent environmental requirements must be addressed as well as challenges in the form of new, more complex minerals. Product development of new types of blast furnaces and direct reduction pellets continues. LKAB’s logistics system from mines and railways to ports also must be overhauled. The next generation of locomotives and wagons are automated, have higher energy efficiency and can handle an increased axle load. Double tracks on the Ore Line are also a necessity. To successfully meet the above challenges and ambitions, the correct working methodology must be applied and the value of new technologies and methods must be fully utilized.
Picture: LKAB
8
MBE OR DBE? Considering LKAB’s large future investments and the need for extensive system development, the question arises of which working methodology is best suited to achieve the desired goals in an efficient manner. The total investment cost of up to SEK 400 billion clearly indicates that extensive system analyses are required to minimize risks and incorrect choices along with the ambition to develop world-unique, new and relatively untested technology. A method for describing how the choice of document- or model-based working methods affects a system’s lifecycle cost has been developed (Madni & Purohit, 2019). Here, the following three key parameters are identified as decisive for which methodology should be used: 1. System complexity 2. Environmental complexity 3. System lifespan System complexity is based on the number of components, the interplay between components, the need for skills to develop the system, and the amount of information required to describe the system. Environmental complexity is defined by the number of stakeholders, regulations, and applicable standards and requirements. The third parameter, system lifespan, is considered low if the lifespan is 0 to 1 year, while a lifespan of 30 years or more is considered high. By assessing and classifying all three key parameters as “low”, “medium” or “high” for a number of different industry segments, the industries most likely to benefit from the implementation of MBSE could be identified. For the industrial segment “Natural Resources” (LKAB), all three key parameters were classified as “high”, which is in concurrence with the general perception within LKAB. Analysis with the described method shows that it is probably favorable to implement and use MBE-MBSE at LKAB.
9
A model-based approach often requires a greater resource investment in early project phases than a document-based approach (Madni & Purohit, 2019). This depends on the initial work required to define requirements, establish models and verify and validate the requirements and models. The merit of the MBE comes from the expanded opportunity for early error detection, optimization, risk minimization, reuse and improved communication. In particular, this merit is the ability to detect errors and deviations from optimal production that can lead to significant cost savings later in the system lifecycle. Overall, the value of MBSE increases as the degree of complexity increases.
Comparison of costs between traditional and model-based engineering (Madni & Purohit, 2019)
10
• MBE uses models for design and analysis before the physical facility is built or updated.
• With DBE, models are used for limited analysis and/or design, sometimes after a facility has been built.
Principle comparisons between document- and model-based engineering methods
11
Get started with MBE A number of criteria and work areas have been identified to clarify the direction of travel toward a model-based approach. The proposals vary in complexity, objectives and time spent and should not be considered a complete compilation. MBE will clearly influence organization and work methods and is part of the general ongoing digitization. Among companies that conduct project-based activities, a common way to distribute responsibilities involves the project management controlling what is to be done and when it is to be done, while the line management determines who is to execute the work and how it is to be performed. This principle can also be applied to the process of steering a business toward a more model-based approach. Line management can ensure that long-term and strategically correct work methods are used and are responsible for competence building in the area. The line organization is also suitable for managing results, data and models over time. Implementing and using the described principles avoids a situation where different projects use different working methods and tools and resources are not used optimally. In principle, important preconditions for achieving a more model-based approach are the following: • Management has approved the direction of travel and initiates relevant activities • Line management controls the choice of tools and working methods ( how ) • Line management works with competence building ( who ) • Results, data and models are managed in a long-term and structured manner
12
MODEL USE An effective way to ensure the quality of development projects is to use models early in the process for both functional analysis and visualization in 2D and 3D. Models are also necessary to enable advanced process control, production planning and maintenance. Below are a number of areas where models can be used that are of particular interest for LKAB to develop. In some of the areas, there are ongoing initiatives that require further development and additional focus to form a complete MBE mindset. • Development projects Model-based engineering is initiated early in relevant development projects where intelligent digital descriptions of process flows and component data in the PDM/ALIM-product data management/asset lifecycle information management system are the foundation. Interoperability of generated data enables efficient project work and supports subsequent project phases. Structures and models are created, reused, and refined in subsequent lifecycles. Dynamic system models are established based on produced static descriptions and describe overall key metrics such as production capacity, economy, environment and energy use. Concept comparison is developed for prioritization between the alternatives. Requirements for system performance are developed and form a well- worked basis for the procurement of components and systems. • Construction work The Building Information Model (BIM) is the model of the facility or building itself, and Virtual Design and Construction (VDC) is a way of working that ensures that all participants in a project have access to models in the right way and at the right time. VDC is a systematized way to make models accessible and value-creating throughout the construction process. It connects all actors both at the level of detail and on the project as a whole. • Production Physical dynamic models that follow the process in realtime provide opportunities to use the models’ predictive abilities at any time. Control algorithms, production planning, action analysis and training are made possible in an efficient and distributed manner in the facilities. Insights into nonmeasurable conditions, so- called soft sensors, are provided through model-based estimation.
13
• Maintenance By comparing the outcome between a model and reality, anomalies can be detected and identified and measures initiated. Models are an excellent method for error detection. Predictive maintenance is made possible by models that describe wear mechanisms which are then used to predict the time of an accident.
• Management Models within management are used for the following: - Dissemination of information - Basis for policy-making - Education - Production planning - Overall business management
Superior models create insights into key metrics that are used to manage resources and efforts in the most effective way from plant- and site-wide perspectives. With effective production planning tools, for example, the consequences of rock outbursts that lead to replanning in the mine can be tested virtually to see the complete short- and long-term system and customer impacts. Optimal alternative plans can be simulated and identified, for example, cross-driving of goods between plants and shafts.
14
CONCEPT • IDEA • PILOT STUDY • DESIGN FORECAST - Concept and global dynamical models - Concept presented in P & ID diagrams - Concept is priorities - fundamental 3D-documents - Selected concept refines and specifies dimensions - VR is used to establish and inform - Requirements are defined
IMPLEMENT • DESIGN • BUILD • CONSTRUCT • OPERATE - BIM and VDC - Development of control - Code verification and acceptance test - Global models are iterated based on current demands - Training of operators - Detailed models for design and verification
PRODUCE • MAINTENANCE • TERMINATE - Detection of anomalies - Verify plans and pinpoint measures -
Reusable education for critical operation Process-based real-time simulation
- - -
Model based maintenance
Key figures and condition estimation
Model usage in different key stages
15
METHOD DEVELOPMENT
• MBE methodology General MBE methodology must be developed and documented. Important elements include identifying and defining common models (both descriptive and predictive) with the aim of avoiding duplication of work and increasing quality. Project manuals can be updated to clearly steer project work toward a model- based approach. Training efforts such as webinars for relevant staff are also necessary to establish the methodology. • Interoperability Interoperability can be defined as “the ability of different systems, often in computer contexts, to work together and be able to communicate with each other”. Important standards in the area are ISO 15926 and CFIHOS. The SEIIA organization conducts standardization work in Sweden, but there is a need for industry representatives to participate in the work to drive development in a direction that is appropriate for current operations. • Automatic/semiautomatic model generation There are several reasons to replace manual model generation with automatic or semiautomatic model generation. Risks of errors and deviations are reduced and a more consistent person-independent work process is obtained. Predictive models can, for example, be generated from intelligent P&ID schemas with linked descriptive models and data (i.e., ALIM systems and CAD models). This means that development focus and resources can be directed toward work with superior process functionality and not toward subordinate code generation. • Generation and updating of models Complex facilities and systems are constantly changing. New requirements and new technology mean that outdated and less suitable technical solutions and components are replaced. This means that connected models may need to be updated to accurately reflect the corresponding physical facility. Since this work can be resource intensive, methodology is needed to automate the process of creating, calibrating and integrating models in a real-time environment and keeping them current.
16
• Modular approach When it is possible to divide systems, models and information into modules, a number of advantages can be obtained, including the following: - Facilitates handling of complex processes - Simplifies reuse, change and formation of new combinations - Simplifies parallel work - Simplifies collaboration with suppliers and external partners - Promotes creativity and collaboration International standards are also available for the division of computer programs into modules (FMIs). This methodology adds value as platform- independent programming as well as the ability to create “blackbox modules”, i.e., modules where the code is not readable by external persons. This may be suitable for collaboration with, for example, external partners and suppliers. • Models with a control module The development of control system models should move toward implementing code in high-level languages rather than in strongly hardware-connected vendor-specific software. This methodology has been used in the automotive industry for the past ten years, where today, a majority of the models are developed in this way, driven by shorter development time, higher quality requirements and more available development resources. The method also enables a more flexible and distributed functional development where process models and control models can be executed together. Ongoing activities within IEC 61131–10, where among other things, the PLC open XML format is specified, are enabling this modularization development.
17
TOOLS AND INFRASTRUCTRE
• Common development and management environment • A common environment must be defined for the development and
management of models and related information. Aspects to consider include access and rights management, development tools, version management, interoperability and storage. The main objectives are to create transparency and facilitate practical work and handling.
• Visualization tools • Virtual reality (VR) is computer technology that simulates real or virtual environments and our presence and interaction in them. Augmented reality (AR) is a live viewing of a physical, lifelike environment whose elements are amplified by computer-generated sensory impressions such as audio, video, graphics, or GPS data. For example, you can experience VR/AR with special glasses that show realtime images of a computer-based image/model of the environment. Both technologies (VR and AR) have a number of uses within MBE, such as visualization in development work, construction, virtual meetings, training and maintenance. FURTHER READING ABOUT MBE: INCOSE: incose.org SEIIA: seiia.se
VINNOVA AutoTwin: vinnova.se/p/autotwin-pre Data Exchange in the Process Industry: dexpi.org
Madni, A.M., Purohit, S., Economic Analysis of Model-Based Systems Engineering Systems, 2019, 7(1), 12: doi.org/10.3390/systems7010012
18
19
20
Page 1 Page 2 Page 3 Page 4 Page 5 Page 6 Page 7 Page 8 Page 9 Page 10 Page 11 Page 12 Page 13 Page 14 Page 15 Page 16 Page 17 Page 18 Page 19 Page 20Made with FlippingBook Ebook Creator