The creation of a DigitalTwin knowledge graph data model confronts the need for access to measurement data in order that the DigitalTwin can create timely performance metrics, identify promptly performance issues, and so on.

However, the quantity of raw data in an Industrial IoT is staggering. A typical process manufacturing plant might have greater than 100,000 measurement points each of which is streaming data by the second or even faster. So how can the raw data be integrated to allow performance analysis?

Just … important😓: 

  • Someone(?) determines a subset of the data that can and should be replicated. 
  • A problem is that it is inevitable someone will ask a yet unformulated question about data that has *not* been replicated. Can you afford to keep revising your ETL?

Just … in case😧: 

  • The bullet is bitten and all data is replicated. 
  • An issue is the DigitalTwin now has to resolve the problem of handling vast quantities of data that has already been solved by IIoT applications. Why not stick to solving unsolved problems?

Just … reporting😫: 

  • Someone writes a dedicated application that pulls data from the DigitalTwin and the IIoT just when reporting. 
  • A problem is that every new performance metric or calculation requires another dedicated application, even if that same metric has been already created in another application.

Just … in time😂: 

  • The KnowledgeGraph pulls the IIoT data whenever required by an IntelligentGraph calculation. 
  • There are no limits on what IIoT data can be requested, but no storage or replication issues either. Also, the IntelligentGraph calculations can access the results of other IntelligentGraph calculations, simplifying the deployment of metrics.

See this short video demonstration of how easy an #IoT-connected #DigitalTwin of a process plant #IntelligentGraph can be created:

The IntelligentGraph-way or the Groundhog-way to efficient data analytics.pdf

Data is rather like poor red wine: it neither travels nor ages well. IntelligentGraph avoids data traveling by moving analysis into the knowledge graph rather than moving data to the analysis engine, obsoleting the groundhog-analysis-way

Solving data analysis, the IntelligentGraph-way

Data is streamed to the IntelligentGraph datastore, and then analysis/calculation nodes are added to that graph which is accessible to all and by all applications.

The IntelligentGraph-way of data-analysis is to:

  • Extract-Load (losslessly) the source data into an Intelligent Knowledge Graph
  • Add analysis/calculation nodes to the KnowledgeGraph which calculate additional analysis values, aggregations, etc.
  • Report results, using any standard reporting tool
  • Add more analysis/calculation nodes as additional requests come through. These new calculations can refer to the existing results
  • … relax, you’ve become the star data analyst:-)

Solving data analysis, the Groundhog-way

Data is in the operational data-sources, data is staged in a data-warehouse/mart/lake, then the analysis is done by the analysis engine (aka Excel), right? And like poor red wine, constantly moving data damages it.

The Groundhog-way of data analysis is to:

  • Extract-Transform(aka probably damage the data as well)-Load the source data into a data-warehouse/mart/lake just to make it ‘easier’ to access.
  • Realizing the required analytical results are not in the data-warehouse/mart/lake, extract some data into Excel/PowerBI/BI-tool-of-choice where you can write your analysis calculations, aggregations, etc.
  • Report analysis results, but forget to or cannot put the analysis results back into the data-warehouse/mart/lake.
  • Repeat the same process every time there is a similar (or identical) analysis required.
  • … don’t relax, another analysis request shortly follows 🙁

IntelligentGraph Benefits

IntelligentGraph moves analysis into the knowledge graph rather than moving data to the analysis engine, avoiding the groundhog-analysis-way:

  • Improves analyst performance and efficiency
    • Eliminates the need for analysts to create ELT to move data to the analysis engine. 
  • Simplifies complex calculations and aggregations
    • PathQL language greatly simplifies navigating and aggregating throughout the graph.
  • Ensures calculation and KPI concurrency
    • Calculations are performed in-situ with the data, so no need to re-export data to the analysis engine to view updated results.
  • Uses familiar scripting language
    • Scripts expressed in any of multiple scripting languages including Python, Javascript, Groovy, and Java.
  • Improves analysis performance and efficiency
    • Time-to-answer reduced or eliminated as analysis is equivalent to reporting
  • Ensures analysis effort is shared with all
    • Analysis results become part of the graph which can be used by others ensuring consistency.
  • Self-documenting analysis path to raw data
    • The IntelligentGraph contains calculation scripts that define which calculations will be performed on what data (or other calculation results).
  • Improves analysis accuracy by providing provenance of all calculations
    • Trace of any analysis through to raw data is automatically available.
  • Simplifies reporting
    • Reporting tools can be used that focus on report appearance rather than calculation capability since the latter is performed in the IntelligentGraph. 
  • Highly scalable
    • IntelligentGraph is built upon the de-facto graph standard RDF4J, allowing for the use of any RDF4J compliant datastore.
  • Standard support
    • Access to IntelligentGraph for querying, reporting, and more is unchanged from any RDF-based KnowledgeGraph.
  • Evolutionary, not revolutionary modeling and analysis
    • Graph-based models offer the ability to evolve as data analysis needs grow, such as adding new dimensions to the data, unlike a ‘traditional’ data mart or warehouse which usually require a rebuild.
  • Creates the Intelligent Internet of Things
    • Scripts can access external data, such as IoT, on-demand allowing IoT-based calculations and analysis to be performed in-situ.
  • Eliminates spreadsheet-hell
    • All spreadsheet calculations and aggregations can be moved into the graph, leaving the spreadsheet as a presentation tool. This eliminates the problem of undocumented calculations and inconsistent calculations in different spreadsheets.

IntelligentGraph embeds calculation and analysis capability within RDF knowledge graphs, rather than forcing analysis to be undertaken by exporting query results to external analysis applications such as Excel.

IntelligentGraph achieves this by embedding scripts into the RDF knowledge graph which are evaluated when queried with SPARQL.

Scripts, written in a variety of languages such as JavaScript, Java, and Python, access the underlying graph using simple pathPatternQL navigation.

IntelligentGraph=KnowledgeGraph+Embedded Analysis.pdf

Why IntelligentGraph?

At present calculations over stored data are either delivered by custom code or exporting the stored data to spreadsheets. The data behind these tools is inevitably tabular. In fact, so dominant are spreadsheets for analysis that the spreadsheet itself becomes the ‘database’ with the inherent difficulties of syncing that data with the source system of record.

The real-world is better represented as a network or graph of interconnected things Therefore a knowledge graph is a far better storage organization than tables or objects. However, there is still the need to perform ad hoc numerical analysis over this data. 

RDF DataCube can help organize data for analysis, but still the analysis has to be performed externally. Confronted with this dilemma, knowledge graph data would typically be exported in tabular form to a datamart or directly into, yet again, a spreadsheet where the analysis could be performed.

IntelligentGraph turns this approach on its head by embedding the calculations as scripts within the knowledge graph. These scripts are evaluated on query, and utilise the data in situ: no concurrency issues.. This allows the calculations to have knowledge of its neighbouring nodes and edges, just like Excel cells can access other cells in the spreadsheet. Access to other nodes within the graph uses pathPatternQL navigation.

Example Data and Analysis

An Industrial Internet of Things (IIoT) application is connecting all the measurements about a process plant, such as an oil refinery, into a knowledge graph that relates the measurements to the material flows through the process equipment.

Although there is an abundance of measurements and laboratory analyses available, the values required for operating and performance monitoring are not (and mostly cannot) be directly measured. 

For example:

  • Stream Mass-Flow: direct mass flow measurements are rare. Instead, a volume flow measurement is used in conjunction with a measured material density to calculate the mass-flow
  • Unit Mass Flow Throughput: this is calculated by summing either all feed stream mass flows or product stream mass flows.
  • Unit Mass Balance: this is calculated by differencing the feed from product mass flows
  • Product Stream Yield: this is the ratio of a stream’s mass-flow to the unit to which the stream is connected throughput.

Figure 1: Typical Process Flow Sheet

These are simple examples; however, they show the reliance on the knowledge graph structure to perform the analysis.

Solving data analysis, the traditional way

Data is in the database, analysis is done by the analysis engine (aka Excel), right?

Figure 2: Data analysis the traditional way

In this scenario, the local power user sets up a query to export data from the database and converts to a format that can be imported into Excel. Ever increasingly complex formulae are then written to wrangle the data into the results that are required.

Why is the spreadsheet approach risky?

  • The analysis is now separated from the data. Data changes will not be reflected in the analysis. Worse still, changes to the analysis might not be propagated to all the spawned copies of the spreadsheet.
  • The data is separated from the analysis. The analysis results are rarely re-imported into the data store where data vs analysis could be performed. Instead, even more data is extracted into the spreadsheet.
  • The difficultly of managing the separation of data from analysis becomes so great that in many cases the database is dispensed with entirely and the spreadsheet becomes the de-facto database.

Solving data analysis with an IntelligentGraph

The beauty of Excel is that a cell can contain either a value or a formula that can reference other cell’s values. Why not do the same with a graph: a node can have edges that terminate with a literal value, or a formula that can reference other node’s values.

This is illustrated in the diagram below:

  • The :massFlow property is not measured directly, so a formula is used for its value instead. This formula references $this, the node to which the calculation is attached, and uses the method getFact() to retrieve related values. The argument of getFact() is a pathPatternQL expression.
  • The :totalProduction property is not measured directly, so a formula is used instead which iterates over all of the ‘stream out’ nodes, retrieving the value of the :massFlow for each stream. The :massFlow value is, of course, in turn a calculation.

Figure 3: Intelligent Graph Data Analysis

Why is the IntelligentGraph approach so advantageous?

  • There is no separation between data and analysis, removing the risk of stale and inaccurate data and calculations.
  • The calculations embedded within the graph can take advantage of the knowledge that is contained within that graph. This makes the calculations far simpler than those that need to be embedded in spreadsheets.
  • The calculations will automatically utilize on the fly the changing knowledge.

How does IntelligentGraph Work?

Analysis is embedded in an IntelligentGraph simply by adding script literals as object values of subjects with datatype of the scripting language (groovy, javascript, python etc).

The IntelligentGraph engine is provided as an RDF4J Stackable SAIL. This means that its capabilities can be combined with any other RDF4J capabilities. The choice of RDF storage remains the same as for any other RDF4J compliant framework.

Modeling with Scripts

Typically, a graph node will have associated attributes with values, such as a stream with volumeFlow and density values:

Stream Attributes:

   :density ".36"^^xsd:float ;
   :volumeFlow "40"^^xsd:float .

Of course, in the ‘real-world’ these measured values are sourced from outside the KnowledgeGraph and change over time. IntelligentGraph can deal with both of these requirements.

The ‘model’ of the streams can be captured as edges associated with the Unit:

   :hasProductStream :Stream_1 ;
   :hasProductStream :Stream_2 ;

Calculate Mass Flow

The calculations are declared as literals[1] with a datatype whose local name corresponds to one of the installed script languages:

:Stream_1 :massFlow     
     _this.getFact(':volumeFlow');"^^:groovy .

Calculate Total Production

A typical performance metric is to understand the total production from a unit, which is not of course directly measured. However, it can be easily expressed using existing calculated values:

:Unit_1   :totalProduction
    "var totalProduction =0.0;
    for(Resource stream : _this.getFacts(':hasProductStream'))
        totalProduction += stream.getFact(':massFlow');
    return totalProduction; "^^:groovy .

Instead of returning the object literal value (aka the script), the IntelligentGraph will return the result value for the script.

We can write this script even more succinctly using the expressive power PathQL:

:Unit_1  :totalProduction  
    "return _this.getFacts(':hasProductStream/:massFlow').total(); "^^:groovy

However, IntelligentGraph allows us to build upon existing calculations to simply express what would normally be difficult-to-calculate metrics, such as product yield or mass balance.

Calculate Mass Yield

Any production unit has different valued products. So a key metric is the yield of individual streams. This can easily be calculated as follows, using values that are themselves calculations.

var result= _this.getFact(":massFlow").floatValue()/ 

Calculate Mass Balance

Measurements are not perfect, nor is the operation of a unit. One of the first indicators of a problem is when the mass flow in does not match the mass flow out. This can be expressed as another calculated property of a Unit:

return  _this.getFacts(":hasFeedStream/massFlow").total() -_this.getFacts(":totalProduction").total();

Querying Results

Access to the calculated values is via standard-SPARQL. However instead of returning the script literal, IntelligentGraph will invoke the script engine, 

Thus to access the :massFlow calculated value, the SPARQL is simply:

select ?massFlow
 :Stream_1 :massFlow ?massFlow 

If the script literal is required then the object variable can be postfixed with _SCRIPT:

select ?massFlow ?massFlow_SCRIPT 
 :Stream_1 :massFlow ?massFlow, ?massFlow_SCRIPT 

If a full trace of the calculation, including tracing calls to other scripts, is required then the object variable can be postfixed with _TRACE:

select ?massFlow ?massFlow_TRACE 
 :Stream_1 :massFlow ?massFlow, ?massFlow_TRACE 

How to Write IntelligentGraph Scripts?

Script Languages

Any Java 9 supported language can be used simply by making the corresponding language JAR available. 

By default, JavaScript, Groovy, Python JAR are installed. The complete list of compliant languages is as follows

AWK, BeanShell, ejs, FreeMarker, Groovy, Jaskell, Java, JavaScript, JavaScript (Web Browser), Jelly, JEP, Jexl, jst, JudoScript, JUEL, OGNL, Pnuts, Python, Ruby, Scheme, Sleep, Tcl, Velocity, XPath, XSLT, JavaFX Script, ABCL, AppleScript, Bex script, OCaml Scripting Project, PHP, Python, Smalltalk, CajuScript, MathEclipse

Script Context Variables

In addition, each script has access to the following predefined variables that allow the script to access the context within which it is being run.

  • _this, a Thing corresponding to the subject of the triples for which the script is the object.  Since this available, helper functions are provided to navigate edges to or from this ‘thing’ below:
  • _property, a Thing corresponding to the predicate or property of the triples for which the script is the object.
  • _customQueryOptions, a HashMap<String, Value> of name/value pairs corresponding to the pairs of additional arguments to the SPARQL extension function. These are useful for passing application-specific parameters.
  • _builder, a RDF4J graph builder object allowing a graph to be constructed (and manipulated) within the script. A graph cannot be returned from a SPARQL function. However the IRI of the graph can be returned, and any graph created by a script will be persisted.
  • _tripleSource, the RDF4J TripleSource to which the subject, predicate, triple belongs.

Fact and Path Functions

The spreadsheets’ secret sauce is the ability of a cell formula to access values of other cells, either individually or as a set. The IntelligentGraph provides this functionality with several methods associated with Thing, which are applicable to the _this Thing initiated for each script with the subject Thing.

Thing.getFact(String pathPattern) returns Value

Returns the value of node referenced by the pathPattern, for example “:volumeFlow” returns the object value of the :volumeFlow edge relative to _this node. The pathPattern allows for more complex path navigation.

Thing.getFacts(String pathPattern) returns Values

Returns the values of nodes referenced by the pathPattern, for example “:hasProductStream” returns an iterator for all object values of the :hasProductStream edge relative to _this node. The pathPattern allows for more complex path navigation.

Thing.getPath(String pathQL) returns Path

Returns the first (shortest)  path referenced by the pathQL, for example “:parent{1..5}” returns the path to the first ancestor of _this node. The pathQL allows for more complex path navigation.

Thing.getPaths(String pathQL) returns PathResults

Returns all paths referenced by the pathQL, for example :parent{1..5}” returns an iterator, starting with the shortest path,  for all paths to the ancestors of _this node. The pathQL allows for more complex path navigation.

Path Patterns

Spreadsheets are not limited to accessing just adjacent cells; neither is the IntelligentGraph. PathPatterns provide a powerful way of navigating from one Thing node to another. PathPatterns are inspired by SPARQL and propertyPaths, but a richer, more expressive, PathQL was required for the IntelligentGraph.


Examples of PathQL patterns are as follows:


will return the first parent of $this.


will return the first child of $this.


will return the grandparents of $this.


will return the siblings of $this.

_this.getFacts(“:hasParent[:gender :female]/:hasParent”)

will return the maternal grandparents of $this

_this.getFacts(“:hasParent[:gender :female]/:hasParent[:gender :male]”)

will return the maternal grandfather of $this.

_this.getFacts(“:hasParent[:gender [ rdfs:label “female”]]”)

will return the mother of $this but using the label instead of the IRI.

_this.getFacts(“:hasParent[eq :Peter]/:hasParent[:gender :male]”)

will return the grandfather of $this, who is the parent of :Peter.

_this.getFacts(“:hasParent[ne :Peter]/:hasParent[:gender :male]”)

will return grandfathers of $this, who are not the parent of :Peter.

The following diagram visualizes a path through a genealogical graph, from _this to the find the parents of a maternal grandfather born in Maidstone:

_this.getFacts(“/:parent[:gender :female]/:parent[:gender :male, :birthplace [rdfs:label ‘Maidstone’]]/:parent”)


Figure 4: PathPatternQL Example

How Is Performance?

IntelligentGraph takes the following actions to improve performance:

  1. All intermediate calculation results are cached, keyed by the subjectNode, predicate, and customQueryOptions.
  2. Cache can be cleared using the SPARQL function ClearCache.
  3. The SPARQL function ObjectValue takes as its argument the subject, predicate and objectValue. If the objectValue supplied is not of script datatype, the function will immediately return the objectValue.
  4. Circular functions, in which A calls B calls A, are detected and rejected.

Can I Debug?

Since IntelligentGraph combines calculations with the knowledge graph, it is inevitable that any evaluation will involve calls to values of other nodes which are in turn calculations. For this reason,  IntelligenrtGraph supports tracing and debugging:


Figure 4: Tracing Calculation

How Do I Add Intelligence to my RDFGraph?


The project is located in Github, from where the intelligentgraph.jar can be downloaded from there:

The intelligentgraph.jar does not include all of the scripting etc language dependencies, so to use it you would have to be certain all dependencies are already available.


IntelligentGraphwill work only with RDF4J version 3.3.0 and above. 

Copy intelligentgraph.jar

To  /usr/local/tomcat/webapps/rdf4j-server/WEB-INF/lib/intelligentgraph.jar

The RDF4J server will need to be restarted for it to recognize this new JAR and initiate the scripting engine.

[1] In this case the script uses Groovy, but any Java 9 compliant scripting language can be used, such as JavaScript, Python, Ruby, and many more.