Dejan Gradišar

Production-process modelling based on production-management data: a Petri-net approach

During the development of a production control system, an appropriate model of the production process is needed to evaluate the various control strategies. This paper describes how to apply timed Petri nets and existing production data to the modelling of production systems. Information concerning the structure of a production facility and the products that can be produced is usually given in production-data management systems. We describe a method for using these data to construct a Petri-net model algorithmically. The timed Petri-net simulator, which was constructed in Matlab, is also described. This simulator makes it possible to introduce heuristics, and, in this way, various production scenarios can be evaluated. To demonstrate the applicability of our approach, we applied it to a scheduling problem in the production of furniture fittings.

IEC 61131-3 Compliant Control Code Generation from Discrete Event Models

This paper describes a control logic implementation approach, which is based on discrete event models in form of finite state machines and Petri nets. Such models may be derived during supervisory control synthesis. The approach defines a transformation of the models into an IEC 61131-3 compliant code that can be translated and downloaded into a standard industrial programmable logic controller. This way, the development and implementation phases of industrial automation projects are shortened significantly. A well proven solution libraries may be built by developed and tested models and reused when necessary

Production modelling for holistic production control

Abstract Holistic production control is a concept that introduces production optimisation by employing model-based, closed-loop control of the principal production Performance Indicators (pPIs). The concept relies on the development of a simple black-box model that describes the relation between the main pPIs and the most influential input (manipulative) variables. In this article the modelling aspects of the holistic production control implementation are presented. The main steps of the production modelling procedure are described, such as data preprocessing, the definition of pPIs, the selection of input variables and the derivation of black-box models. Particular emphasis is given to a modelling approach based on neural networks and a corresponding modelling assistant tool, which has been developed to support the modelling procedure. The approach is illustrated on the Tennessee Eastman benchmark process, where neural network models for three main production performance indicators, i.e., costs, quality and production rate, are derived.

Closed-loop control of a polymerisation plant using production performance indicators (PIs)

The synthesis of plant-wide control structures is recognised as one of the most important production-management design problems in the process industries. This article proposes a closed-loop control structure utilising production performance indicators (pPIs) as a possible solution to simplify this problem. pPIs represent the translation of operating objectives, such as the minimisation of production costs, to a set of measurable variables that can then be used in a feedback control. The idea of closed-loop control at the production-management level using pPIs as referenced controlled variables was implemented on the procedural model of a production process for a polymerisation plant, and two types of controllers were tested: an experimental controller based on look-up tables, and an advanced model predictive controller (MPC). Preliminary results show the usefulness of the proposed methodology.

Automated Petri-net modelling based on production management data

Timed Petri nets can be used for the modelling and analysis of a wide range of concurrent discrete-event systems, e.g. production systems. The present paper describes how to do so while starting from the information about the structure of a production facility and about the products usually given in production-data management systems. We describe a method for using these data to algorithmically build a Petri-net model. The timed Petri-net simulator, which was built in Matlab®, is also described. This simulator makes it possible to introduce heuristics, and in this way production operations can be scheduled. To demonstrate the applicability of our approach, we applied it to a scheduling problem in a multi-product batch plant.

ProOpter: An advanced platform for production analysis and optimization

Abstract This paper presents the prototype of an advanced platform for production analysis and optimization, referred to as ProOpter. The platform was developed to support the recently derived concept of holistic production control (HPC), which relies on model-based control. The prototype is comprised of a set of off-line and on-line modules. The off-line modules support the definition of key performance indicators (KPIs), the selection of the most influential input (manipulative) variables, and the identification of a simple production model from historical data. The on-line modules enable KPI prediction and suggest actions to the production manager, employing model-based production control and/or optimization techniques. In this way, a new decision-support reasoning based on historical production data can be introduced. ProOpter has a modular design and can be used as an add-on to existing production IT systems since it relies on established industrial communication standards. The use of the platform is validated on the well-known Tennessee Eastman benchmark simulation process and on two industrial case studies.

Automated Petri-Net Modelling for Batch Production Scheduling

The control of batch processes poses difficult issues as these processes are neither continuous nor discrete, but have the characteristics of both. ISA society introduced a multi-part S88 standard where the first part [1] defines the models and terminology for batch plants and control systems. S88 provides a framework for the development of technologies that not only support control activities of batch processes but also management activities, such as scheduling. This is illustrated in [14] where a generic framework is defined for interpreting a multi-purpose/product batch plant in terms of S88 constructs for scheduling purposes.

Input variable selection algorithms for HPC

The paper refer to the problem of a holistic production control (HPC). The main idea of HPC is to derive an optimisation strategy which is based on a simple model of only a few production Key Performance Indicators (pKPI). The main challenge of the HPC approach is the derivation of an appropriate pKPI model, where the main steps are: data preprocessing, pKPI definition, input variables selection (IVS) and black-box modelling. In this paper we are focused on the problem of the selection of the most influential inputs, where the review of different IVS methods is given.

Production Control of a Polymerization Plant using a Reduced Set of Control Variables

The specifics of process manufacturing have a great influence on production management, and the focus of process-production control is to maintain stable and cost-effective production within given constraints. The synthesis of production-control structures is thus recognized as one of the most important design problems in process-production management. This paper proposes a closed-loop control structure with the utilization of production-performance indicators (pPIs) as a possible solution to this problem. pPIs represent the translation of operating objectives, such as the minimization of production costs, to a reduced set of control variables that can then be used in a feedback control. The idea of production-feedback control using production PIs as referenced, controlled variables was implemented on a procedural model of a production process for a polymerization plant. Some preliminary results demonstrate the usefulness of the proposed methodology.

Controllability of holistic production control

Holistic production control introduces a concept for production process optimisation, which is based on detailed analysis of historical process data. Selection of the most influential manipulative variables represents an important HPC design step. Therefore, the selected set of variables should be additionally validated with controllability analysis. Appropriate space-based controllability measure for HPC is presented, where achievable output space is examined. Although controllability measure is based on the non-optimal regression model, it is assumed that such examination of the model knowledge still gives us some additional insight into process and eases our input selection procedure. The HPC controllability approach is illustrated on the Tennessee Eastman benchmark process.