STEP-SF4.0 is a research and development project that is a public-private partnership between the South African Government's Department of Science and Innovation's (DSI) Sector Innovation Fund (SIF) programme and the South African sugarcane processing industry. The SMRI's proposal for the project was approved by the DSI on 17 January 2019 and is overseen by a Steering Committee comprising government and industry representatives and a Technical Review Committee comprising of industry experts.
Historically, the South African sugar industry has attained very high levels of efficiency and process performance, supported by generations of specialist sugar technologists and operational staff who had invested their entire careers in the industry, many of whom have either retired or relocated. Whilst it is acknowledged there are several reasons for the marked decrease in process performance and efficiency that has occurred at least over the last decade, the loss of key skills and knowledge has certainly contributed significantly to this. For example, the loss of pan boiling skills is an often-cited example that has resulted in a decreasing trend in boiling house recovery. There is a great need to capture and embed the knowledge of industry experts and skilled operation staff in interactive operating procedures, decision-support and troubleshooting toolkits and control systems, and improve the performance and competitiveness of the South Africa sugar processing industry.
The project, entitled “STEP-SF4.0 – Sugarcane Technology Enabling Project for Sugar Factory 4.0" has been designed to develop and implement information management systems and decision-support tools aligned with the 4th Industrial Revolution/Industry 4.0 smart manufacturing to enable the sugar processing industry to make meaningful improvements over current factory performance levels. The theme of the STEP-SF4.0 project is to combine sugar technology, data analytics and smart manufacturing principles to drive productivity and efficiency in sugar factories. Sub-projects have been constructed to target areas of the sugar factory where significant improvement potential has been identified, and especially where these build on existing tools and projects that have shown promise.
Industry 4.0 is a name increasingly applied to initiatives that arise from or lead to increased digitalisation of industries, targeting process improvement, productivity optimisation and automation. These objectives are made possible by recent advances in:
Industry 4.0 projects can range in scope from modules that focus on single unit operations to factory or industry-wide solutions for improving productivity and efficiency. Industry 4.0 applications can involve any, or all, of the nine digital industrial technologies viz. 1) big data analytics, 2) simulation, 3) additive manufacturing, 4) horizontal and vertical data/information integration, 5) industrial internet of things, 6) augmented reality, 7) advanced robotics, 8) cloud computing and cybersecurity.
Figure 1 provides a visualisation of the architecture of an Industry 4.0 application in industry. Each application must be driven by factory operating data, i.e. local data sources. These are typically housed in information management systems such as the factory data and laboratory information historians and offline systems including equipment installation manuals, spreadsheets and paper-based records. During the development phase of each application, historical performance data may be interrogated via empirical, mechanistic or simulation analysis to identify causal and correlated agents between available measured factors and calculated and measured plant performance. The learnings from the development phase are synthesised and compiled into a deployment application that is intended to be run live in the factory. The deployment application should collect and analyse incoming data in real or near-real time and provide outputs that facilitate rapid and effective decision-making or trouble shooting. The deployment application outputs must integrate with a visualisation system (i.e. trending software, custom dashboard, interactive simulator or “digital twin") that provides the right information to the right people, especially those responsible for decision-making, corrective action and overall production oversight, at the right time.
Figure 1: Architecture of Industry 4.0 projects (adapted from Wallner PHF (2018) How Industry 4.0 is changing our way of developing. Seminar: Future proof your business for the 4th Industrial Revolution. 21 June 2018 Umhlanga)https://smri.sharepoint.com/:b:/s/teams/ResearchDevelopment/EW3gBY4WLfpJphL89Q4iWaABmo9EqSSqj0snIbHMzX-lKg?e=CBpd5U
The overall application must be built on infrastructure that allows data collection, aggregation, analysis, storage, retrieval and transmission. This may be locally based at the factory, but could also be remotely based, e.g. in the factory administration offices or even in the head office of companies with more than one technical operation. In both cases hardware and communication protocols will be needed to permit the required level of data management.
A significant feature of Industry 4.0 projects is that the software and hardware for each application can be chosen uniquely to suit the needs of the factory and the application. The Development phase portrayed in Figure 1 utilises MATLAB®, StatisticaTM and Microsoft® Excel. Any other platforms that allow the necessary data aggregation and analysis could be used instead. Similarly, the deployment application could be built in any of several suitable applications. The visualisation step involves development of dashboards and graphical user interfaces that allow factory staff to easily access the knowledge that the Industry 4.0 modules create. This is where the potential of any application is translated into actual benefits. Visualisation can be implemented on existing factory systems or in custom-built applications using software involved in development, deployment¸ or third-party visualisation software.
Figure 2 is a common representation of the stages of Industry 4.0. Industry 4.0 projects may deliver information on the status of a process (stage 1), analysis of why the process has the observed status (stage 2) and predictions on the progression of the process status (stage 3). Stage 4 projects use the combined information from the previous stages to develop automation and control technologies. Each one of these stages has potential to create value opportunities for a particular focus area, through highlighting areas of loss or cost, identifying causes of lost opportunity, providing strategies for reducing lost opportunity, or controlling the process to a higher level of efficiency. The value of benefit will vary between applications and will depend on the efficiency attained before their implementation. As a general rule, the value of the benefit achieved is anticipated to be higher at each successive stage of implementation. Thus, any project that can generate an implementable solution at any one of the four stages is expected to result in cost reduction or increased value recovery.
The STEP-SF4.0 project is targeting the first three stages. Automation projects are not anticipated to be tackled within the programme, but recommendations for potential automation applications may be among the outputs of the programme.
Figure 2: Staged approach to industry 4.0 (adapted from https://www.i-scoop.eu/industry-4-0/#Industry_40_design_principles)
While SMRI staff generally lack the operations skills for running factories efficiently, they are uniquely positioned to facilitate the development of Industry 4.0 applications. The key knowledge investment into these applications will take place in the development and deployment phases. These will require the supervisory role to be carried out by individuals who have a thorough knowledge of sugar technology, who employ systems thinking, have sound understanding of statistics in analysis of large complex data sets with strong non-linearities and time dependency, have strong process modelling skills, and have sufficient time to investigate different strategies to achieve a defined goal. For example, these skills are currently being employed by the SMRI researchers working on the STEP-Bio Energy Monitoring project, which is being expanded upon in this STEP-SF4.0 project.
In addition, the SMRI is the custodian of an enormous database of factory performance figures, dating back many decades, and it has strong relationships with all sugar factories in South Africa and their affiliate mills in southern Africa. Also, the SMRI is the primary developer of NIRS prediction equations for the rapid analysis of intermediate sugar factory streams. Offering this as a base technology into the STEP-SF4.0 project enables the almost real-time analysis of fructose, glucose and sucrose of all sugar process streams – a world first - and enables the production of new factory performance data and insights to be gained. The SMRI has a strong track record of collaboration with higher education institutions in South Africa and internationally and has historically leveraged these collaborations to bring new ideas and technologies into the South African sugar industry.
It is acknowledged that, to remain globally competitive, individual factories or sugar companies might be considering implementing Industry 4.0 projects within their own factories using in-house technical support staff and consultants. However, the nature of these projects is somewhat different to traditional sugar technology projects; Pioneer Industry 4.0 projects around the world make use of advanced data analytics skills and software application development. There are few, if any, local or international experts or consultants who have experience integrating these skills with sugar technology. An important objective of the proposed project is to draw on the wisdom of skilled sugar factory and SMRI staff and the capabilities of advances in data analytics to codify sugar technology knowledge into applications that facilitate improvements in factory performance metrics.
A project scoping workshop was held at the SMRI in December 2018 to inform the industry of the project and the concept of Sugar Factory 4.0 and to generate ideas for projects that could be undertaken within the time frame. These ideas were developed into clearly defined subprojects and calls were issued for sugar factories and consultants in the 4IR field to submit proposals for Technical and Value Assessments (TVAs) that would identify solutions to listed problems and the costs and benefits of piloting the solutions in factories. Six TVAs were approved involving five projects, three factories and four consultants. These TVAs were undertaken during 2019 and the final reports were considered by the TRC and the Steering Committee to approve which projects should proceed to the piloting phase. Four piloting projects were approved as follows to commence in May 2020:
Host Factory: Illovo Sugar (South Africa) Pty Ltd, Noodsberg Sugar Mill
Consultant: Mzukulu Technologies (Pty) Ltd
The overall project aims and expected outcomes are a reduction in the quantity of sucrose lost to final molasses in a sugar factory. The objectives are to: (1) investigate the causes of excessive sucrose losses to final molasses at the C-centrifuge station, using SMRI-NIRS and other technologies, and (2) investigate means of employing data analysis technology to monitor sucrose losses to molasses in real time coupled with an assessment of possible causative factors in the boiling house to provide actionable insights to operational staff.
Host Factory: RCL Foods Sugar & Milling (Pty) Ltd, Komati Sugar Mill
The proposed solution will collect all sources of data on the diffuser and use machine learning to investigate relationships between observed and controllable variables and the diffuser performance in terms of extraction. If the information content of measured variables is enough for the machine learning model to be able to predict diffuser performance, it will be able to identify conditions that have a positive or negative impact on diffuser performance trajectory.
Consultant: Opti-Num (Pty) Ltd
Objective: Effective management of factory throughput in a sugar factory.
The consultant has considered the factory as a sequence of sub-sections with intermediate flow buffering. A throughput bottleneck is a section which cannot generate output at the rate that the subsequent section can process received input. The proposed solution will analyse variables correlated with the development of bottlenecks to identify opportunities to pre-empt or mitigate throughput problems.
Host Factory: Gledhow Sugar Company (Pty) Ltd
Consultant: Stone Three Digital (Pty) Ltd
Overall project aims: Reduction of energy consumption
Objectives are to monitor:
The scope of the project is to replicate the existing SMRI steam supply and vapour bleed consumption monitor in a custom platform and to develop a series of dashboards to allow factory staff to interact with the monitor.