The need for quality improvement and accountability in food supply chains-particularly concerning outcomes like hunger-demands a comprehensive approach that integrates product quality with its impact on stakeholders and communities. A Traceable Impact Management Model (TIMM) provides a structured framework for evaluating stakeholder roles across production and consumption, enhancing traceability's role in assessing community-level impacts. Aligned with regulatory demands and stakeholder needs, TIMM is grounded in an ontological model that ensures consistent logic and terminology. This integrated solution fosters global traceability, promoting sustainability, accountability, and responsible food systems on a broader scale. With these combined eforts, the food supply chain moves toward a global tracking and tracing process that not only ensures product quality but also addresses its impact on a broader scale, fostering accountability, sustainability, and responsible food production and consumption.
United Nations Sustainability Goal 2, “Zero Hunger,” redirects attention from production metrics to the food supply chain’s impact on stakeholders experiencing hunger and insecurity1. Achieving food security demands inclusive strategies—especially for vulnerable populations—emphasizing traceability of how supply chain activities influence access to safe, nutritious, and high-quality food.
Traceable impact in the agricultural food chain involves modeling production, processing, distribution,
and consumption to assess socio-economic and environmental outcomes [
This paper introduces an ontology-based framework integrating impact management and traceability, grounded in ISO/IEC 5087-1 and the Common Impact Data Standard (CIDS). It presents the Traceable Impact Management ontology as a foundation for modeling, assessing, and improving stakeholder outcomes in the food supply chain.
Traceability in the food supply chain entails tracking and retracing each step of food production,
processing, transportation, and consumption, including who performed activities, when and where they
occurred, and the quantity and quality involved [
Beef production is impacted by environmental, biological, and socio-economic factors, including disease, worker conditions, and logistics. Cross-jurisdictional transport introduces risks of spoilage, contamination, and regulatory incompatibility, with economic variables like pricing and distance influencing access [ 5, 6]. Final delivery often reveals cumulative food insecurity risks such as expired or nutritionally compromised products.
Traceability fosters accountability from raw material sourcing to consumption (Figure 2). It enhances quality control, enables eficient recalls, supports regulatory compliance, and eliminates redundant inspections by sharing accurate upstream data. The food supply chain includes agriculture, agroindustries, trade/distribution, and supporting industries. Stakeholders span producers, processors, logistics, consolidators, and consumers.
Modeling traceability involves methodologies like Petri-nets, IDEF, EPC, UML, BPM, SADT, EDOC, and ADF [7, 8, 9, 10, 11, 12, 13]. These visualize and analyze workflows to improve traceability and operational eficiency. Efective modeling distinguishes planned networks from actual events, requiring data on actors, products, equipment, and transitions, captured using RFID, NFC, GS1, and EPC standards.
Traceability exists at two levels: internal (within organizations) and global (across systems). Internal traceability tracks detailed product data, inputs, equipment, and storage. Global traceability connects stages across stakeholders, focusing on broader data like origin, transport, and animal identification. Both levels are essential for securing transparent, accountable, and sustainable food systems.
Impact encompasses the intended and unintended changes that afect organizations and their stakeholders over time as a result of activities2. In the food sector, such changes include outcomes influenced by production, labeling, and claims governed by relevant policies3. Despite varying frameworks like Logic Models or Theory of Change, impact models share core elements that enable unified representation [14]. The Impact Management Project4 defines impact across six dimensions: What, Who, How Much, Contribution, Risk, and How. CIDS incorporates these to assess outcome nature, afected stakeholders, extent of change, and associated risks [15].
Within the food supply chain, impact management links outcomes to traceable events and resources, from sourcing to consumption. This approach accounts for food waste, safety, and sustainability as essential outputs of supply chain activities [16]. Traceability enhances stakeholder understanding of how supply changes can drive community benefits, such as improved environmental practices or food security. Visibility into raw materials motivates producers to maintain quality, boosting trust and accountability.
Further, traceable impact systems support timely interventions, like targeted recalls, when product issues afect downstream stakeholders. Identifying afected communities—especially distant or indirect ones—enhances response efectiveness and minimizes harm. These systems also improve quality control by linking material origins to community impacts.
Strategically, implementing traceability and impact management ensures regulatory compliance and market adaptability. It supports sustainability mandates, enables diferentiation through transparent practices, and strengthens consumer confidence. As policy and consumer expectations evolve, these systems become vital for ethical, competitive, and resilient food supply chains.
This section reviews ontologies that underpin the Traceable Impact Management ontology: ISO/IEC 5087 and TOVE for representing enterprise activity and traceability; and CIDS for impact modeling. 3.1. TOVE and 5087 The ISO/IEC 5087-1:2023 standard, rooted in TOVE [17, 18], defines core concepts like activities, resources, time, agents, and organizational structures. Its Activities-State model identifies how actions emerge from enabling states and produce caused states. The TOVE-based activity abstraction further structures activities into sub- and super-activities, capturing their logical and temporal dependencies through clustered states, enabling a systematic trace of processes such as “produce ground beef.”
Temporal modelling in 5087-1, based on OWL-Time [19, 20, 21], provides constructs for representing time-points, intervals, and temporal relations. These enable detailed scheduling and sequencing of activities and states, allowing planners to evaluate overlaps and precedence relationships for optimized coordination.
The 5087-1 resource ontology defines resources through quantity, unit, time, location, and allocation. It assesses divisibility and utility, supporting decisions on scheduling and impact tracing [22]. For example, splitting a batch of ground beef into multiple units contrasts with the indivisibility of a single patty box. Beyond raw materials, the ontology encompasses machines, energy, and labor, highlighting their social, environmental, and economic impacts throughout the supply chain—from production to transportation and consumption. 2https://innoweave.ca/en/modules/impact-measurement 3https://inspection.canada.ca/en/food-labels/labelling/industry/method-production-claims 4https://www.theimpactprogramme.org.uk/portfolio/impact-management-project/
TOVE’s Traceability Ontology [23] formalizes quality through logical constructs, emphasizing conformance to requirements via sub-domains like measurement and traceability. The Traceable Resource Unit (TRU) defines resources at a specicfi time and place, quantifying them at the point of interaction with primitive activities. TRUs maintain fixed quantities post-creation and cannot be subdivided internally. Aggregation of multiple TRUs forfeits individual identity, necessitating careful handling for trace accuracy. Trace paths [24] connect TRUs and activities via :PrimitiveTrace instances, enabling a networked trace across the supply chain that supports granular quality and impact analysis.
The Common Impact Data Standard (CIDS) [15], developed under the Common Approach to Impact Measurement5, provides a standardized yet flexible framework for modeling and evaluating the impact of organizational activities. Initially aimed at social purpose organizations (SPOs), it addresses the challenges of consistent measurement and reporting across varying domains. Supported by multiple community partners and government funding, CIDS enhances the harmonization of impact measurement.
Standardization through CIDS enables unified data collection, supporting optimized impact delivery, complex analyses (e.g., longitudinal studies), and adaptable reporting for diferent stakeholders. It reduces administrative overhead by ofering formats that meet diverse funder’s requirements, enhances content discoverability online, and promotes integration with global standards like the UN SDGs, IRIS+, and IATI.
CIDS models impact using core classes: Organization (entity responsible for impact), Stakeholders (individuals or groups afected), Characteristic (codes identifying stakeholders), Activity (actions producing or enabling outcomes), Output (quantified activity results), Outcome (stakeholder experiences), Indicators (metrics with time, location, value, and units), Indicator Report (metric value over time), Impact Report (captures scale, depth, duration of outcomes), and Impact Risk (likelihood and materiality of deviation in impact). It integrates ISO/IEC 5087-1 ontologies for activities, resources, and time, enabling traceable and standardized impact representation across systems.
This section defines the extensions to existing ontologies that focus on traceability of resources across changing states, claims made, and related stakeholders.
The :Resource pattern defines the relationship between the resource being discussed and the activity that changed its state and a stakeholder related to that activity, as demonstrated in Figure 3. The :Resource is a subset of Manifestation, extending it with three properties, :hasQuantity with amount and units of measure, :existsAt with its temporal dimension, and :hasLocation with its geospatial dimension. The resource is related to a :TerminalState as part of activities that change its state, which can be one of :ProduceState, :ConsumeState, :ReleaseState, or :UseState. For example the initial production of, say “beef” would be defined as a ProduceState, while consumption of that “beef” by the customer would be defined as a :ConsumeState.
To ensure resources are traceable, they are tracked using the :PrimitiveTrace class. Each :PrimitiveTrace class relates the resource at one point in time to another point with :traceFrom and :traceTo properties. Each :PrimitiveTrace is part of a continuous :Trace comprised of one or more individual :PrimitiveTrace instances, as demonstrated in Figure 4. It should be noted that :traceTo and :traceFrom properties for a :Resource are separate from :inputOf and :hasOutput properties that may be found in process ontologies. Rather, :traceTo and :traceFrom relate the :Resource to the :trace, not the activities that may have produced or used it.
The :Claim class identifies what claims are being made on a resource, such as “locally produced” and “organic.” The :Policy class represents how those claims are defined. For example, given the :Claim of “locally produced,” a :Policy can be defined that applies to the claim if and only if geospatial distance between the locations where the resource was produced (meat packer) and consumed (customer) is less than some predefined threshold, say 1,000km. Depending on the policy, diferent Resource properties can be referenced using the :satisfiedBy property. For example, in Figure 5 we see that the distance calculation relies on the two :Resource manifestations at times 0 and 1. The :Policy axiom can then be defined to reference properties as needed, namely the location of the resource. The trace pattern allows for the traceability of claims as the state of the resource changes, as demonstrated in Figure 5 . The :Claim is related to the :PrimitiveTrace the claim is for with the :claimFor property.
In order to align the claims with outcomes that stakeholders are interested in, such as “eating locally produced food,” the :Claim class is related to the CIDS :Indicator class. An :Indicator is used to report on whether the claim is true or not, given the state of the resource. The :Indicator reports on the boolean values of “true” or “false” indicating whether the claim applies or not, respectively, as illustrated in Figure 6. As long as the Policy applies to the claim, the :Indicator’s value can be reported as True.
We align the 5087-1 Activity and Resource ontology with TOVE’s traceability ontology and CIDS indicators to construct a model for tracing impact. While CIDS provides a comprehensive model for representing activities and impact, TRUs provide a model to connect activities and resources at time points and locations. The proposed model simplifies the representation by: 1) replacing a :TRU with a 5087-1:Resource, 2) replacing the “curp_tru” classes with corresponding 5087-1:TerminalState subclasses, such as “produce_tru” with :ProduceState, and so on, 3) associate the :PrimitiveTrace class with the :Resource directly instead of the :TRU, and 4) assign the cids:Indicator with the :Claim class.
This allows us to assign :TRU properties that may change over time, namely quantity, time, and location to the :Resource class directly. Figure 7 illustrates a complete traceable resource and claim. Due to limitations of description logic-based ontologies, first-order-logic axioms are used to specify the semantics of the concepts. For example, the applicability of policy on a claim, such as Equation 1, requires rules that are not supported by description logic.
We demonstrate the model’s usage in the next section with a scenario outlined from Figure 8 to Figure 12. Here, we are interested in the claim that food delivered to consumers is produced locally, where the definition of locally produced goods is anything that was produced within 1000km from the place of use or consumption. At each stage, the outcome stakeholders are interested in reporting on is whether food product is locally produced.
To determine whether a :Resource, say “Beef,” was used or consumed within 1,000 km of production, we rely on at least one :Policy that defines what “locally produced” means, at least one :Claim on that :Resource in question, an :Indicator that can be reported, and the :Outcome the stakeholder is interested in, as per Figure 7. As long as the :Policy applies to the :Claim, the :Indicator’s value can be reported as True. The :Resource “Beef” has a manifestation :mbeef0, to represent the existence of “beef” at time 0. It is linked to the :PrimitiveTrace which connects it to the next step in the trace, namely “Pack Produced Beef.” This :PrimitiveTrace is part of the larger :Trace instance that will be extended in the next section.
Equation 1 provides an example of a :Policy axiom, which validates that the geospatial distance between a resource’s first manifestation and its current position is less than or equal to some threshold value, say 1000km.
∀ , ∃, ∃ [ ℎ ℎℎ(, ) ∃0, ∃0, ∃0, ∃ , ∃ { ℎ ( , 0)∧ ℎ(0, 0) ∧ : (0, 0) ∧ _(0, 0)∧ ℎ( , ) ∧ : ( , ) ∧ _( , )∧ 72 : ( ) ∧ 72 : (0) ∧ 72 : ( )∧ (( ≥ 0) ∧ ( − 0 ≤ )) } ⊃ (, , ) (1) where • : Resource being evaluated at the current space and time • 0, : Locations of resource at the point of its first manifestation and its current ( ) location • 0, : Quantities representing the distance between the location where the resource was produced and its current location • : Quantity representing the threshold below which a resource is considered locally produced. • : Policy that defines what constitutes a locally produced resource.
The calculation is based on the latitude and longitude of the locations being compared, namely the location of the first manifestation at L0 and the current location at LC.
In Equation 2, we give an example of a :Claim axiom for the LocallyProducedClaim. This axiom defines what resource R and policy P the claim C is for.
∀, ∃, ∃ [ (, , ) ≡ where • : Claim representing the impact claim • : Resource the impact claim is made about • : A terminal state for resource • : Policy that defines what constitutes an organic resource
Finally, in Equation 3, we give an example of an :Indicator and its claim, namely the :LocallyProducedIndicator I, its corresponding truth value M, and its claim C. • : Resource being evaluated at the current space and time • : Claim being evaluated • : Policy that defines what constitutes a locally produced resource. • : Indicator used to measure whether a claim is true or not • : Measure used to store the value of indicator , where the value type is a boolean
In the following scenario, a meat packing plant in Guelph, ON produces 200 lbs of beef, which is transported 664 km to a grocery store in Montreal, QC where it is purchased by consumers. As shown in Figure 8, the organization “Westcoast Meat” is an instance of class :MeatPacker, and the CIDS class :Organization. It performs the activity of producing beef (“Beef Production”), which is an instance of the 5087-1:Activity class. At time 0 the organization performs the “Beef Production” activity which produces 200 lbs of beef at their plant in “Guelph, ON.” The activity has the efect of creating the (2) (3) :ProduceState at the time point 0, signifying the creation of 200 lbs of beef at location “Plant, Guelph, ON.” At this point, the “beef” becomes an instance of a :Manifestation, superclass of :Resource. While trivially true, illustrates the representation of the “LocallyProduced” :Claim for this Resource. The :Resource has the location “Plant, Guelph, ON” as the location of production.
Figure 8 also illustrates the next activity and changes in the :Resource, namely the packing of produced beef onto a truck. The packing of the beef onto the truck, “Pack Beef”, is an instance of the :ConsumeState class since it consumes the existing resource, 200 lbs of beef, at time point 1. The location has changed from “Plant, Guelph ON” to “Truck, Guelph, ON.” This example illustrates the linking of two :Resources using the :PrimitiveTrace class, which links a :Resource to another. In Figure 8, the “Pack Produced Beef on Truck” :PrimitiveTrace is a trace from “Beef Production” :Activity to the next activity, “Pack to Truck.”
According to the LocallyProducedPolicyApplies axioms in Equation 1, the locations of activities “Produced Beef” and “Pack Beef” are used to calculate the diference, with a :Measure value of “0.1 km”. Given that the beef was packed onto a truck 0.1 km away, it also satisfies the claim that the beef is locally produced, and the “LocallyProduced” Indicator has a value of “True.”
Figure 9 illustrates the next activity’s :Resource, namely the transport of the produced beef. The organization “Bob’s Trucking” is an instance of both the :Trucker class and :Organization class. It transports beef from “Truck, Guelph, Ontario” to its final destination, namely “Truck, Montreal, Quebec.” The “Trucked Beef” :UseState transition’s the beef’s state from “Produced Beef” to “Trucked Beef” since it uses the resource beef without consuming any of it, creating a new :Resource at time-point 2. Given that the transport’s destination is within 1000 km, it also satisfies the claim that the beef is locally produced, and the “LocallyProducedClaim” has a value of “True.”
Figure 10 illustrates the next two steps in the supply chain, namely the unloading and stocking the beef to the grocery store 644 km from the meat packer. At time point 3, the grocery store organization “Joe’s Supermarket” receives the beef and unloads it in their store. “Joe’s Supermarket” is an instance of both the :GroceyStore class and :Organization class. It performs the “Unload Beef” activity which “releases” it from the trucks possession of the beef. Next, it performs the activity of “Stock Beef”, creating the :ProduceState at 4 with 200 lbs of beef. Again, since the location of the “Stocked Beef” :Activity is within 1000 km of the meat packer, this beef is locally produced, and the “LocallyProducedClaim” is “True” at time-point 4.
The final step in the supply chain is the selling of beef to consumers. Figure 11 illustrates this by tracing the buying of 2 lbs of beef by a consumer at time point 5. The consumer “Mary”, is an instance of the :Consumer class as well as a member of the CIDS class :Stakeholder that identifies a group of stakeholders that are interested in buying locally produced food. The consumer performs the activity of “Buy Beef” in the amount of 2 lbs at time-point 5. This activity transitions a portion of the 200 lbs of beef, namely 2 lbs, from “Stocked Beef” to “Bought Beef,” creating a :ConsumeState :tru at time point 5 for 2 lbs of the beef shipment.
In Figure 8, the “Locally Produced Policy X” is calculated according to Equation 1 from the point of production to point of consumption by the customer. The locations of “Produced Beef” :Resource (“Plant, Guelph, ON") and “Bought Beef” :Resource (“Home, Montreal, QC”) are used to calculate the diference indicator with a value of “644 km”. Given that the beef was bought 644 km away from being produced, it satisfies the claim that the beef is locally produced, and the “LocallyProduced” indicator has a value of “True” at 5. Hence, the “LocallyProducedClaim” is true. By tracing the claims across all primitive traces that comprise the main “Deliver beef to customer” :Trace, the claims remain true throughout.
Finally, in Figure 13, we align the newly calculated “LocallyProduced” indicator with the outcome consumer “Mary” is interested in. The consumer’s :Activity instance “Buy Beef” has an efect, namely the :ConsumeState of “Bought Beef.” An instance of the :StakeholderOutcome class represents the outcome the consumer is interested in, namely “Buy Local Food.” The indicator “LocallyProduced” is then linked to the stakeholder outcome through the hasIndicator property. The LocallyProducedPolicyApplies axiom ensures that the claim and Indicator are true. Again, the “Locally Produced” calculation is based on the calculated distance “644 km” and threshold of “1000 km”, giving “True” for the “LocallyProduced” indicator value.
The proposed model outlines an ontological framework for the representation and mechanics of how to integrate impact modeling, activity modeling, and traceability. The model focuses on the transformation of resources as they move through a supply chain. The resources are tracked through four diferent types of changes imposed by an activity, namely production, use, consumption, and release. The claim class allows one to place a label on the state of the resource, while a policy allows one to define how the claim is determined. Separating claims and policies in this way allows us to define one label for multiple policies, ensuring that a claim can be defined in more than one way. By representing the changing state of the resource as part of a trace, we can track what activities occurred during various stages of the resource, moving backwards and forward in time, evaluating the claim’s applicability at each point. Finally, by extending the claim pattern with CIDS, we can associate each claim with a stakeholder and a desired outcome, relying on the impact indicator to report on whether the claim is true at any given point in time during the trace.
The provided scenario illustrates the model’s ability to trace the claim of “local production” as it pertains to the production, transport, sale, and consumption of beef. The model could be extended beyond sales and identify other impacts on the consumer and beyond, such as the environment. For example, the model can trace how a consumer’s nutrition levels are impacted. With the use of counterfactuals, we can identify the risk of, say, not delivering beef to grocery stores on the consumer’s nutritional intake of protein and other nutrients. In our scenario, 200 lbs of beef would impact 100 consumers in the community. This shortfall must be reconciled by other means. Not doing so would pose a risk to the healthcare system by potentially causing malnutrition in parts of the population.
Equally, we can trace the environmental risk of food supply chain. For example, consider that the beef is not transported properly above, say 4.4C, and must be thrown out. While the consumer is impacted by not consuming the beef, the environment is impacted by producing food waste, adding 60,000 kg of carbon dioxide and contributing to global warming. Meanwhile, the rotten beef and packs need to be dealt with separately due to diferent material properties. Rotten beef will be incinerated and continue to produce greenhouse gases, while the pack of plastic may go to a landfill, especially in some developing countries, thus creating further damage to the soil structure.
This paper presents the Traceable Impact Management Model (TIMM), an ontology-based framework for aligning food supply chain traceability with societal and environmental impact claims. Built upon ISO/IEC 21972, 5087-1, and the Common Impact Data Standard, TIMM enables a comprehensive connection between product quality and broader stakeholder outcomes, addressing hunger and sustainability while meeting evolving regulatory and consumer demands.
At its core, the model defines a unified terminology to represent stakeholders, resources, and their transformation across production, use, consumption, and release. A claim-policy structure allows for evaluating the state and trajectory of resources, while impact indicators provide validation of stakeholder outcomes [15]. By tracing resource history and associating claims with specific actors and objectives, the model supports proactive assessments and actionable recommendations to improve systemic accountability across the supply chain.
TIMM ofers a coherent logic for modeling data artifacts and quality processes, enabling stakeholders to visualize goods’ transformations. Its shared vocabulary ensures consistent understanding and coordination. Additionally, this ontological framework provides the foundation for developing software systems that capture and analyze data for traceability, supporting quality assurance, recalls, and regulatory compliance. Through this integration, TIMM fosters a globally traceable, accountable, and sustainable food system.
The author(s) used Grammarly to spell-check and ChatGPT-4o to reduce length of some paragraphs in this work. Authors reviewed and edited the final publication, and take full responsibility for its content. [5] M. Carney, Compounding crises of economic recession and food insecurity: A comparative study of three low-income communities in Santa Barbara County, Agriculture and Human Values 29 (2012) 185–201. doi:10.1007/s10460-011-9333-y. [6] A. C. Hoek, S. Malekpour, R. Raven, E. Court, E. Byrne, Towards environmentally sustainable food systems: decision-making factors in sustainable food production and consumption, Sustainable Production and Consumption 26 (2021) 610–626. [7] S. Balamurugan, A. Ayyasamy, K. S. Joseph, Enhanced petri nets for traceability of food management using internet of things, Peer-to-Peer Networking and Applications 14 (2021) 30–43. [8] M. Thakur, C. R. Hurburgh, Framework for implementing traceability system in the bulk grain supply chain, Journal of Food Engineering 95 (2009) 617–626. [9] M. Bevilacqua, F. E. Ciarapica, G. Giacchetta, Business process reengineering of a supply chain and a traceability system: A case study, Journal of Food Engineering 93 (2009) 13–22. [10] S. Islam, J. M. Cullen, L. Manning, Visualising food traceability systems: A novel system architecture for mapping material and information flow, Trends in Food Science & Technology 112 (2021) 708–719. [11] S. P. Gayialis, E. P. Kechagias, G. A. Papadopoulos, N. A. Panayiotou, A business process reference model for the development of a wine traceability system, Sustainability 14 (2022) 11687. [12] T. Pizzuti, G. Mirabelli, The global track&trace system for food: General framework and functioning principles, Journal of food Engineering 159 (2015) 16–35. [13] C. Verdouw, A. Beulens, J. Trienekens, J. Wolfert, Process modelling in demand-driven supply chains: A reference model for the fruit industry, Computers and electronics in agriculture 73 (2010) 174–187. [14] K. Ruf, How impact measurement devices act: the performativity of theory of change, sroi and dashboards, Qualitative Research in Accounting & Management 18 (2021) 332–360. [15] M. Fox, K. Ruf, A. Chowdhury, B. Gajderowicz, T. Abdulai, J. Zhang, The Common Impact Data Standard: An Ontology for Representing Impact, Technical Report, Technical Report, University of Toronto, Toronto, Canada, 2021., 2021. URL: https://commonapproach.org/wp-content/uploads/ 2020/12/Common-Impact-Data-Standard-V1.1.pdf. [16] V. S. Yadav, A. Singh, A. Gunasekaran, R. D. Raut, B. E. Narkhede, A systematic literature review of the agro-food supply chain: Challenges, network design, and performance measurement perspectives, Sustainable Production and Consumption 29 (2022) 685–704. [17] M. S. Fox, The tove project towards a common-sense model of the enterprise, in: International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems, Springer, 1992, pp. 25–34. [18] M. S. Fox, J. F. Chionglo, F. G. Fadel, A common-sense model of the enterprise, in: Proceedings of
Industrial Engineering Research Conference, 1993, pp. 178–194. [19] S. Cox, C. Little, Time ontology in owl w3c recommendation 19 october 2017. w3c candidate recommendation, World Wide Web Consortium (W3C) (2017). URL: https://www.w3.org/TR/2022/ CRD-owl-time-20221115/. [20] J. R. Hobbs, F. Pan, Time ontology in owl, w3c working draft, World Wide Web Consortium (W3C) (2006). URL: https://www.w3.org/TR/2006/WD-owl-time-20060927/. [21] J. F. Allen, Towards a general theory of action and time, Artificial intelligence 23 (1984) 123–154. [22] F. G. Fadel, M. S. Fox, M. Gruninger, A resource ontology for enterprise modelling (1994). [23] H. M. Kim, M. S. Fox, M. Gruninger, An ontology of quality for enterprise modelling, in: Proceedings 4th IEEE Workshop on Enabling Technologies: Infrastructure for Collaborative Enterprises (WET ICE’95), IEEE, 1995, pp. 105–116. [24] H. M. Kim, M. S. Fox, M. Grüninger, Ontology for quality management - enabling quality problem identification and tracing, BT Technology Journal 17 (1999) 131–140. doi: 10.1023/A: 1009611528866.