A machine tool management system connects an external server and a large number of NC devices controlling the external server and respective machine tools through a network. The system collects several kinds of signal data from the NC device of each machine tool to the external server. In the system, the external server stores a maintenance period model and a refinement algorithm and obtains a next maintenance period from the maintenance period model.

A machine learning apparatus includes: a state observation unit that observes a characteristic shape, an adopted plan, and a determination result as state variables, the characteristic shape representing a shape of a part of a product of wire electric discharge machining, adjustment of machining conditions being deemed as necessary for the part of the product, the adopted plan being an adjustment method selected from among one or more adjustment methods for adjusting the machining conditions to improve machining performance for the part indicated by the characteristic shape, the determination result indicating whether implementation of the adopted plan is effective in improving machining performance for the part corresponding to the characteristic shape; and a learning unit that learns the machining condition adjustment method according to a data set created based on the state variables.

A machine learning apparatus includes: a state observation unit that observes a characteristic shape, an adopted plan, and a determination result as state variables, the characteristic shape representing a shape of a part of a product of wire electric discharge machining, adjustment of machining conditions being deemed as necessary for the part of the product, the adopted plan being an adjustment method selected from among one or more adjustment methods for adjusting the machining conditions to improve machining performance for the part indicated by the characteristic shape, the determination result indicating whether implementation of the adopted plan is effective in improving machining performance for the part corresponding to the characteristic shape; and a learning unit that learns the machining condition adjustment method according to a data set created based on the state variables.

In the field of process automation there exists one or more custom solutions which are not scalable and missing essential needs to automate a process. This disclosure addresses the problem of codifying a process to automate by a declarative approach. A method for declarative modeling of process for automation is provided. The method includes receiving, request for automating the process; extracting, at least one feature associated with the process to obtain a declarative model; validating, the declarative model of the process with a process automation code to obtain a process automation package; provisioning, the declarative model to access and obtain specific roles associated with the process; executing, the process automation package in a process automation platform based on the specific roles to obtain a plurality of states; and monitoring, the plurality of states and triggers at least one action associated with the plurality of states in the process automation platform.

In the field of process automation there exists one or more custom solutions which are not scalable and missing essential needs to automate a process. This disclosure addresses the problem of codifying a process to automate by a declarative approach. A method for declarative modeling of process for automation is provided. The method includes receiving, request for automating the process; extracting, at least one feature associated with the process to obtain a declarative model; validating, the declarative model of the process with a process automation code to obtain a process automation package; provisioning, the declarative model to access and obtain specific roles associated with the process; executing, the process automation package in a process automation platform based on the specific roles to obtain a plurality of states; and monitoring, the plurality of states and triggers at least one action associated with the plurality of states in the process automation platform.

A machine tool management system connects an external server and a large number of NC devices controlling the external server and respective machine tools through a network. The system collects several kinds of signal data from the NC device of each machine tool to the external server. In the system, the external server collects an estimated thermal distortion from the NC device and compares the estimated thermal distortion with pre-stored data to determine whether the tool is machining favorably.

A method for determining locations of supporting points for a machine tool is proposed. Multiple test coordinate sets in relation to the machine tool are used in cooperation with corresponding deformation index values for multiple components of the machine tool to obtain a deformation index equation that relates to assessed deformation of the components in response to different coordinates for the supporting points. The deformation index equation is used to acquire an optimal coordinate set for the supporting points.

A method for determining locations of supporting points for a machine tool is proposed. Multiple test coordinate sets in relation to the machine tool are used in cooperation with corresponding deformation index values for multiple components of the machine tool to obtain a deformation index equation that relates to assessed deformation of the components in response to different coordinates for the supporting points. The deformation index equation is used to acquire an optimal coordinate set for the supporting points.

A numerical control system includes a computer aided design (CAD) data storage means for storing CAD data, an input/output (I/O) assignment data storage means for storing I/O assignment data, and a relevant information storage means for recording relevant information between the I/O assignment data and mounting information included in CAD data of each of the I/O units, and displays a defect occurrence region on a shape image of a control panel.

A method for stipulating a starting time for an update for a control apparatus that controls at least one peripheral component in a process automation installation by producing control outputs is disclosed. The update requires at least one predetermined down time. The invention is based on a suitable starting time for performing an update in order to perform an update in the course of operation of a process automation installation without risk. The method involves detecting a current operating state for the at least one peripheral component, taking the detected operating state as a basis for using a controlled system model of the process to simulate what state trajectory the at least one peripheral component negotiates within the down time for a predetermined static control output, and checking whether the state trajectory lies exclusively within a range of admissible operating states, and, if appropriate, starting the update.