Method and system for collecting data from a data-source into a MOM data warehouse

Abstract

A process and a system collect data from a data-source into a manufacturing operation management (MOM) data warehouse. The data in the MOM data-warehouse are exposed according to a basic data model in which a performance parameter is linked to a basic set of context identifiers for MOM analysis purposes. The data in the data source are exposed according to a source data model in which a source performance parameter is linked to a source set of context identifiers. A data extensor module is provided for processing the data received from the data source to add, upon need, a context identifier linked to the source performance parameter. Whereby the added context identifier is present in the basic set but it is not present in the source set. The data extensor module processing data is received from the data source to obtain augmented data stored in the MOM data warehouse.

Claims

1. A method for collecting data from a data-source into a manufacturing operation management (MOM) data warehouse, wherein data in the MOM data-warehouse are exposed according to a basic data model in which a performance parameter is linked to a basic set of corresponding context identifiers for MOM analysis purposes, the data in the data source are exposed according to a source data model in which a source performance parameter is linked to a source set of corresponding context identifiers, which comprises the following steps of: providing a data extensor module executed by a processor of a computing device, the processor configured by the data extensor module for processing the data received from the data-source in order to add, upon need, a context identifier linked to the source performance parameter, whereby an added context identifier is present in the basic set but the added context identifier is not present in the source set; processing, in the data extensor module, the data received from the data source to obtain augmented data, the augmented data are exposed in an extended data model compatible with the basic data model; the processing step including the further steps of: generating an additional source performance parameter; and estimating a value of the additional source parameter by adapting a value of the source performance parameter to a linked added context identifier. wherein a user chooses which criterion is used for estimating the value of the additional source parameter, the user choosing one of a numeric distribution technique or a weighted distribution technique taking into account another performance parameter; and collecting the augmented data into the MOM data-warehouse.

2. The method according to claim 1, which further comprises configuring the data extensor module using a user interface.

3. The method according to claim 1, wherein the context identifiers of the source set contains at least one of time identifiers or equipment identifiers.

4. A system for collecting data from a data-source into a manufacturing operation management (MOM) data warehouse, wherein data in the MOM data-warehouse are exposed according to a basic data model in which a performance parameter is linked to a basic set of corresponding context identifiers for MOM analysis purposes, the data in the data source are exposed according to a source data model in which a source performance parameter is linked to a source set of corresponding context identifiers, the system comprising: a data extensor module executed by a processor of a computing device, the processor configured by the data extensor module for processing the data received from the data source in order to add, upon need, a context identifier linked to the source performance parameter, whereby an added context identifier is present in the basic set but it is not present in the source set, said said processor configured by the data extensor module for processing the data received from the data source to obtain augmented data, where the augmented data are exposed in an extended data model compatible with the basic data model; and said data extensor module further configuring the processor to: generate an additional source performance parameter; and estimate a value of the additional source parameter by adapting a value of the source performance parameter to a linked added context identifier, wherein a user chooses which criterion is used for estimating the value of the additional source parameter, the user choosing one of a numeric distribution technique or a weighted distribution technique taking into account another performance parameter: means for collecting the augmented data into the MOM data-warehouse.

5. The system according to claim 4, wherein said data extensor module is configurable via a user interface.

6. The system according to claim 4, wherein the context identifiers of the source set contains at least one of time identifiers or equipment identifiers.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING

(1) FIG. 1 is a diagram schematically illustrating a data model of a MOM data warehouse and a data model of a data source according to the prior art; and

(2) FIG. 2 is a diagram schematically illustrating data collection into the MOM data warehouse from the data source according to an example embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

(3) At least some embodiments of the present invention address the above described issue in which a method and a system collect data from a data-source and put the data into a MOM data warehouse.

(4) Referring now to the figures of the drawings in detail and first, particularly to FIG. 2 thereof, there is shown a diagram schematically illustrating data collection from a data source according to an example embodiment of the present invention. The data in the MOM data-warehouse 101 are exposed according to a basic data model 103 where a performance parameter PP_a is linked to a basic set of corresponding context identifiers CI_1, CI_2, CI3 for MOM analysis purposes. The data in the data source 102 are exposed according to a source data model 104 where a source performance parameter PP_b is linked to a source set of corresponding context identifiers CI_1, CI_2.

(5) A data extensor module 201 is provided for processing the data received from the data source 102 in order to add, upon need, a context identifier CI_3 linked to the source performance parameter PP_b, whereby the added context identifier CI_3 is present in the basic set but it is not present in the source set.

(6) The data extensor module 201 processes the data received from the data source 102 to obtain augmented data, where the augmented data are exposed in an extended data model 202 compatible with the basic data model 103. The augmented data are then collected into the data-warehouse 101.

(7) Advantageously, it is possible to collect the performance parameter PP_b of the data source 102 within the MOM data warehouse 101 and link it to the previously missing context identifier CI_3. As used herein the term “augmentation” indicates that data may be extended and processed since some of the original data might be modified to obtain an estimate of them.

(8) For example, assume that the source data come from a given sensor on given equipment where the sensor is measuring temperature over time. The context of the equipment could easily be inferred by knowing the sensor name and its location but the production order information, eventual useful to the user, would then still be missing. Advantageously, with embodiments, it is possible to add the context production order to the measures of the sensor which are time and temperature.

(9) Example of Algorithm Embodiment

(10) In several shop floor systems, at least one context identifier (time or equipment) is usually available for every performance parameter that is imported in the MOM data warehouse. In this example assume that both context identifiers time and equipment are available in the MOM data sources.

(11) An exemplary algorithm embodiment includes the following steps: a) the user selects a new MOM data source; b) the user selects from the available performance parameters and context identifier of the selected MOM data source, the performance parameter and the linked context identifiers relevant for the desired target analysis, in this case: c) the user selects an additional context identifier—available in the basic data model of the MOM data warehouse but not in the data model of the MOM data source; d) the system loads the selected data from the MOM data source and calculates for each time-equipment pair: d1) all the N-tuple listing the N context identifiers, where N is the number of distinct contexts in the MOM data warehouse and the new MOM data source (where N is bigger than 2); and d2) all the M-tuple listing the M measures, using an augmenting processing engine, where M is number of performance parameters in the MOM data warehouse and in the new MOM data source.

(12) The augmenting processing engine calculates the estimated augmented values according to an estimation criterion, preferably selected by the user.

(13) A first example of user estimation criterion may for example be a numeric distribution approach, whereby the augmented values are calculated using a linear distribution technique.

(14) A second example of user estimation criterion may for example be a weighted distribution approach, where by the augmented values are calculated according to the values of another reference performance parameter.

(15) Examples Illustrating the Numeric and Distribution Approaches

(16) Assume that the MOM data warehouse is configured to collect data from a MES system providing as a performance parameter, energy consumption measures contextualized with the production order. With this configuration, conveniently, the user can analyze the energy consumption for each production order, and compare it for each equipment piece in the plant and at any time period, like e.g. shifts and working weeks.

(17) Assume that the production line of this plant is upgraded with a new MOM data source, which could be a hardware or software one, for example providing scraps quantities that have been generated by the equipment in the production line. Assume that this new data source does not contain any information related to the production order but nonetheless the user wishes to know the quantity of scraps produced for a specific production order.

(18) With embodiments, the user is conveniently enabled to import the data from the new data source into the MOM data warehouse and link it to the existing contexts, including the production order which is the context identifier not provided by this data source. Hence, conveniently, with embodiments, the user has the possibility to query the MOM data warehouse and get the scraps quantity for a specific production order as desired.

(19) This case is exemplified with Tables 1-4 illustrated below.

(20) TABLE-US-00001 TABLE 1 existing data in the MOM data warehouse Context Context Context Measure “time” “equipment” “order” “energy consumption” t1 Eq1 OrdYY1 120 t1 Eq1 OrdYY2 80 t1 Eq2 OrdYY2 70 t2 Eq1 OrdYY1 120 t2 Eq2 OrdYY2 90 t2 Eq2 OrdYY3 30

(21) TABLE-US-00002 TABLE 2 data from the new data source Context Context Measure “time” “equipment” “scraps” t1 Eq1 10 t1 Eq2 6 t2 Eq1 8 t2 Eq2 12

(22) TABLE-US-00003 TABLE 3 augmented data collected in the MOM data warehouse with the numeric distribution technique approach. Measure Estimated (Real Context Context Context “energy Measure Measure “time” “equipment” “order” consumption” “scraps” “scraps”) t1 Eq1 OrdYY1 120 5 (10) t1 Eq1 OrdYY2 80 5 (10) t1 Eq2 OrdYY2 70 6  (6) t2 Eq1 OrdYY1 120 8  (8) t2 Eq2 OrdYY2 90 6 (12) t2 Eq2 OrdYY3 30 6 (12)

(23) TABLE-US-00004 TABLE 4 augmented data collected in the MOM data warehouse with the weighted distribution technique approach. Measure Estimated (Real Context Context Context “energy Measure Measure “time” “equipment” “order” consumption” “scraps” “scraps”) t1 Eq1 OrdYY1 120 6 (10) t1 Eq1 OrdYY2 80 4 (10) t1 Eq2 OrdYY2 70 6  (6) t2 Eq1 OrdYY1 120 8  (8) t2 Eq2 OrdYY2 90 9 (12) t2 Eq2 OrdYY3 30 3 (12)

(24) Table 1 shows the exisiting data in the MOM data warehouse. Such data includes the performance parameter, hereby also called measure, “energy consumption” linked to its corresponding context identifiers “time”, “equipment” and “order”. Hence, as illustrated in Table 1, the basic data model of the MOM data warehouse includes three different context identifiers time, equipment and order.

(25) Table 2 shows data stemming from the new MOM data source. Such data includes the “scraps” measure linked to its corresponding context identifiers “time” and “equipment”. As seen in Table 2, the model of this source data does not contain the context identifier “order” that is present in the basic model of the MOM data warehouse.

(26) Table 3 and Table 4 show the augmented data collected in the MOM data warehouse resulting from the processing of the measured source data real “scraps” by the data extensor module in order to obtain estimated “scraps” (column 5) by linked to the added context identifier “order”.

(27) The values of the estimated “scraps” (column 5) of Table 3 are calculated with a numeric distribution technique and the values of the estimated “scraps” (column 5) of Table 4 are calculated with a weighted distribution technique where the energy consumption measured values are used to weigh the real scraps to calculate the estimated scraps to be associated to each production order.

(28) In this first specific example embodiment of Table 3, scraps are equally distributed between the production orders executed on the same equipment. More in details: a) at time t1 for the equipment Eq1 there are two orders executed, OrdYY1 and OrdYY2, the measured scraps 10 (column 6) are devided equally for the two orders in 5 and 5 (see estimated measure scraps, column 5); b) at time t1 for the equipment Eq2, only order OrdYY2 is executed, therefore the measured real “scraps” 6 of column 6 is reported unchanged into the estimated value of column 5; c) at time t2 for the equipment Eq1, only order OrdYY1 is executed, therefore the measured real “scraps” 8 of column 6 is reported unchanged into the estimated value of column 5; d) at time t2 for the equipment Eq2 there are two orders executed, OrdYY2 and OrdYY3, the measured real “scraps” 12 of column 6 are devided equally for the two orders in 6 and 6 (see estimated measure scraps, column 5).

(29) In this second specific example embodiment of Table 4, scraps are distributed between the production orders executed on the same equipment by taking into account the weight provided by the values of the column “energy consumption” (column 4). More in details: a) at time t1 for the equipment Eq1 there are two orders executed, OrdYY1 and OrdYY2, the measured scraps 10 (column 6) are distributed by the weight provided by the energy consumption values of column 6, 120 and 80, resulting then in 6 and 4 of the estimated measure scraps values of column 5; b) at time t1 for the equipment Eq2, only order OrdYY2 is executed, therefore the measured real “scraps” 6 of column 6 is reported unchanged into the estimated value of column 5; c) at time t2 for the equipment Eq1, only order OrdYY1 is executed, therefore the measured real “scraps” 8 of column 6 is reported unchanged into the estimated value of column 5; d) at time t2 for the equipment Eq2 there are two orders executed, OrdYY2 and OrdYY3, the measured scraps 12 (column 6) are distributed by the weight provided by the energy consumption values of column 6, 90 and 30, resulting then in 9 and 3 of the estimated measure scraps values of column 5.

(30) Advantageously, in embodiments, both examples of Tables 3 and 4, the MOM data warehouse stores both the “original” real imported values of column 6 and the “augmented” estimated values of column 5, thus preserving the complete knowledge of the system, so that the user can query the MOM data warehouse and get the original or augmented values, according to his desired target analysis.

(31) Advantageously, in embodiments, the user chooses which criterion, e.g. numeric distribution or weighted distribution with a reference parameter, to be used by the augmenting processing engine for calculating the estimated augmented values.

(32) In embodiments, the skilled in the art will easily appreciates that it is possible to define in a custom manner other types of estimation criteria for the augmenting processing.

(33) Advantageously, in embodiments, when data from new MOM data sources are imported in the MOM data warehouse, estimated measure values are associated to the imported real measure values and additional context identifiers are rendered available for linking which were originally not included in the MOM data source. Advantageously, in embodiments, the missing added context identifiers are added with a configurational approach without the need to write custom code. Hence, the possibility to create a “link” between database context identifiers and added measures is built in the system thanks to a pre-written core augmenting algorithm, without the need to develop a custom program.

(34) In embodiments, the data extensor module is configurable by the user via a User Interface.

(35) In embodiments, the data extensor module comprises an engine that applies the user selections as regards estimation criteria and/or parameters and context indicators to add. In embodiments, the engine is available with the product.

(36) In embodiments, the module understands what are the missing context indicators and proposes them to the users that accepts them or not.