Sap Ag

Sap Se

Jaakob Kind

Jochen Steinbach

Atreju Florian Tauschinsky

Janick Frasch

Robert Meusel

Minji Lee

Michael Otto

Uta Maria Loesch

Some embodiments include reception of a time-series of a respective data value generated by each of a plurality of sensors, calculation of a regression associated with a first sensor of the plurality of sensors based on the received plurality of time-series, the regression being a function of the respective data values of the others of the plurality of data sources, reception of respective data values associated with a time from and generated by each the plurality of respective sensors, determination of a predicted value associated with the time for the first sensor based on the regression associated with the first sensor and on the respective data values associated with the time, comparison of the predicted value with the received value associated with the time and generated by the first sensor, and determination of a value indicating a likelihood of an anomaly based on the comparison.

Sensor data anomaly detection

A method for multimodal failure analysis is provided herein. A multimodal failure analysis request may be received. An asset type may be determined based on the multimodal failure analysis request. Asset records for the asset type may be obtained. The asset records may include data on asset failures across multiple failure modes. A multimodal failure analytical model may be executed based on the asset records. Executing the multimodal failure analytical model may include calculating a distribution of failure intervals over time, probabilities of failure respectively associated with the failure intervals, and intervention scores respectively associated with the failure intervals. An intervention interval and an intervention score associated with the intervention interval may be selected based on the associated probabilities of failure. The selected intervention interval and intervention score may be provided in response to the multimodal failure analysis request.

Quantification of failure using multimodal analysis

A method and a system for fast recursion check and low-level code generation for directed graph. In a directed graph each vertex is connected to at least one other vertex with a directed edge to form an ordered pair of a source vertex and a target vertex. Low-level codes are assigned to the vertices of the directed graph, where for each ordered pair the low-level code of the target vertex is bigger than the low-level code of the source vertex. A new edge is added to connect a first vertex and a second vertex of the graph, where the first vertex is the source vertex and the second vertex is the target vertex of the new ordered pair. The low-level code of the target vertex of the new ordered pair is not bigger than the low-level code of the source vertex. For the new ordered pair and all ordered pairs that descend from the new ordered pair, the low-level codes of the target nodes are changed to values bigger than the low-level codes of the corresponding source vertices, when necessary.

Method and system for recursion check and low-level code generation for directed graph

Computer-implemented methods, and associated computer program products and systems, are disclosed for generating electronic data for use in planning the execution of a manufacturing process. The methods, generally, are used in the structural transformation of execution-level routings for manufacturing processes into planning-level routings. The methods applies various transformation techniques when a planning-level routing includes alternative paths, and different transformation techniques when a planning-level routing includes parallel paths. Planning modes are created in the case of alternative paths, and parallel paths generally do not cause multiple modes to be created. Parallel and alternative paths in a planning-level operation also has impacts on how the a planned duration formula is determined.

Structural transformation of execution-level manufacturing process routings into planning-level routings

Methods, computer program products and systems are disclosed for generating a planned time duration for a manufacturing process operation defined for planning purposes that includes an aggregation of multiple defined execution operations. Electronic information is received that specifies a production demand that requires use of the manufacturing process for fulfillment. A linear approximation is electronically generated of an overall time duration measure for the defined manufacturing process planning operation. The overall time duration measure is a sum of time components comprising at least a calculated measure for a variable time component for the longest of the multiple execution operations for the received production demand, and a fixed time duration measure for one send-ahead quantity for each of the multiple execution operations that is not the longest. The generated overall time duration measure is then returned.

Generating planning-level time and capacity requirement formulas for manufacturing processes

Generating data used in planning the execution of a manufacturing process includes consistency checks that are performed on proposed groupings of manufacturing operations to make sure that the groupings are defined in a way that enables the information to be aggregated for use in planning. To do this, defined groupings are received for multiple execution-level manufacturing operations that are defined in a manufacturing process execution-level electronic data model and that define a process flow to produce a product. In one aspect, for each of the defined groupings, it is determined whether the defined grouping includes multiple occurrences of a process step definition that has multiple different defined alternative steps, and if so, the defined groupings are changed so that each defined grouping of the defined groupings has no more than a single occurrence of a process step definition that has multiple different defined alternative steps.

Consistency checking and repair of manufacturing operation groupings to be aggregated for use in planning

Embodiments of the present invention relate to a method and system for evaluating selection conditions associated with variants of components of a multi-component configurable product for consistency and completeness.

Growing community of inventors

Heidelberg, Germany

Jaakob Kind