Modelling collaborative services: The COSEMO model

Despite the dominance of the service sector in the last decades, there is still a need for a strong foundation on service design and innovation. Little attention has paid on service modelling, particularly in the collaboration context. Collaboration is considered as one of solutions for surviving or sustaining the business in the high competitive atmosphere. Collaborative services require various service providers working together according to agreements between them, along with service consumers, in order to co-produce services. In this paper, we address crucial issues in collaborative services such as collaboration levels, sharing data and processes due to business inter-dependencies between service stakeholders. Afterward, we propose a model for Collaborative Service Modelling, which is able to cover identified issues. We also apply our proposed model to modelling an example of healthcare services in order to illustrate the relevance of our modelling approach to the matter in hand.

💡 Research Summary

The paper begins by highlighting a gap in the literature: while the service sector has expanded dramatically, systematic modeling approaches for services that involve multiple organizations collaborating have received little attention. The authors identify three core challenges in collaborative services: (1) varying levels of collaboration, (2) the need to share data and processes across organizational boundaries, and (3) business inter‑dependencies that create complex contractual and governance requirements. To address these, they propose the COSEMO (Collaborative Service Modeling) framework, a meta‑model that captures the essential elements of multi‑party service ecosystems.

COSEMO defines three principal roles—Provider, Consumer, and Orchestrator—and three fundamental constructs: Service Objects (the tangible or digital assets involved), Processes (the activities that transform or consume those objects), and Data (the information exchanged among participants). The framework distinguishes between “shared data” (information that multiple parties may read or write) and “shared processes” (activities that are jointly executed by two or more parties). Contracts, expressed as Agreement entities, embed Service Level Agreements, security policies, and cost‑sharing rules, thereby governing access rights and responsibilities. The model is expressed using UML class and sequence diagrams, with OCL constraints to enforce consistency (e.g., a provider may only modify a data item if the contract permits it).

A key contribution is the explicit categorization of collaboration levels. Level 1 involves simple information exchange; Level 2 adds partial process integration; Level 3 represents full process and decision‑making integration. Each level dictates the required granularity of data sharing and the depth of process coupling, allowing designers to select an appropriate level based on strategic goals and technical feasibility.

To demonstrate applicability, the authors model a healthcare service scenario involving a hospital, an insurance company, a pharmacy, and patients. The hospital creates a medical record (a Service Object). The insurance company needs portions of that record for claim verification, while the pharmacy requires the prescription details for dispensing medication. In COSEMO, the medical record is marked as shared data, the claim verification and medication dispensing are modeled as shared processes, and the contracts specify that the insurer may view only diagnostic codes, whereas the pharmacy may view only prescription data. This illustrates how COSEMO can enforce data privacy, maintain integrity, and coordinate workflows across organizational boundaries.

The paper argues that COSEMO integrates smoothly with existing standards such as BPMN and SOA, offering a higher‑level abstraction for multi‑organizational service design while remaining extensible: new roles, objects, or processes can be added without redesigning the entire model. The authors also note that the framework supports non‑functional requirements (e.g., performance, security) through the Agreement component, enabling systematic monitoring and compliance checking.

In conclusion, the authors claim that COSEMO provides a rigorous yet flexible foundation for designing, analyzing, and governing collaborative services. It clarifies responsibilities, formalizes data and process sharing, and embeds contractual constraints directly into the service model. However, the validation is limited to a single healthcare case study; further empirical work across domains such as logistics, finance, or smart cities is needed to assess generalizability. Additionally, the paper suggests future research on tool support for automatic model verification, cost‑benefit analysis of different collaboration levels, and strategies for maintaining model evolution in dynamic business environments.