A resource management system for modular plants includes a database providing a module library of semantic modules representing respective modules in a module pool. At least one of the semantic modules includes a semantic description of the respective module, where the semantic description includes abstract data according to a semantic data model, and where the abstract data describes attributes of the respective module not found in a standard description file for the module. Based thereon, the system facilitates automated generation and optimization of module pipelines.

Techniques are presented for implementing predictive models that learn from real-time manufacturing capacity plan performance data to continually improve the accuracy of generated capacity plans. Instances of a plan performance predictive model are generated in response to receiving then-current sets of capacity plan parameters, predicted plan performance parameters, and measured plan performance parameters that correspond to various capacity plans implemented at a manufacturing facility. Modeled plan parameters produced by the continually adjusted instances of the plan performance predictive model are used to generate candidate capacity plans with progressively more accurate predicted plan performance and/or overall behavior. The candidate capacity plans are analyzed at a user device to facilitate selection of a capacity plan to implement at the manufacturing facility.

Method for error detection and for local limitation of a cause of the error in an installation for machining a workpiece which is preferably formed at least in sections from wood, a wood material, and/or a synthetic material, with the installation having several segments. The method comprises the steps: Detecting a status information which relates to a workpiece throughflow in at least two segments of the installation; determining whether there is an error on the basis of the status information; if there is an error, identifying in which of the at least two segments of the installation the error is present for local limitation of the cause of the error; and outputting a signal containing the information regarding which segment the error is in.

Method for error detection and for local limitation of a cause of the error in an installation for machining a workpiece which is preferably formed at least in sections from wood, a wood material, and/or a synthetic material, the installation having several segments, comprising the steps: Detecting a workpiece parameter in at least two segments of the installation; determining whether there is an error on the basis of the detected workpiece parameter; if there is an error, identifying in which of the at least two segments of the installation the error is present for local limitation of the cause of the error; and outputting a signal containing the information regarding which segment the error is in.

Techniques are presented for implementing predictive models that learn from real-time manufacturing capacity plan performance data to continually improve the accuracy of generated capacity plans. Instances of a plan performance predictive model are generated in response to receiving then-current sets of capacity plan parameters, predicted plan performance parameters, and measured plan performance parameters that correspond to various capacity plans implemented at a manufacturing facility. Modeled plan parameters produced by the continually adjusted instances of the plan performance predictive model are used to generate candidate capacity plans with progressively more accurate predicted plan performance and/or overall behavior. The candidate capacity plans are analyzed at a user device to facilitate selection of a capacity plan to implement at the manufacturing facility.