Comparing Document Delivery Models: Centralized vs. Decentralized Systems

Every research library faces the same basic problem: no single collection can hold everything a user might need. A scholar in Guwahati might require a journal article that exists only in a library in Delhi, and a college in a small town may need a rare report held by a national institution. Document Delivery Services (DDS) exist to solve exactly this gap by moving documents from where they are held to where they are needed. But how a country or a network organizes that movement matters enormously. Should one big library carry the load, or should the responsibility be spread across many? The answer to that question is what separates centralized from decentralized systems, and several scholars have built formal models to map the possibilities.

What document delivery service models actually describe 🔗

A DDS model is essentially a blueprint for who supplies documents, where the master collections sit, and how requests flow through the system. At its core, interlibrary loan and document delivery is a collaborative service that draws on local, regional, and national libraries to obtain materials a user’s own library does not hold. The difference between models comes down to one variable: the degree of concentration. In a highly concentrated system, a few libraries hold the bulk of the supply burden. In a dispersed system, many libraries share that role.

This is not just a theoretical distinction. The structure chosen affects turnaround time, cost, reliability, and how easy it is to charge fees for the service. As the IFLA literature on interlending notes, countries that adopted centralized solutions found it far easier to introduce charges, because the system did not depend on libraries reciprocally exchanging traffic. The model, in other words, shapes the economics.

The four national models from the 1980 UNESCO study 🔗

The foundational framework comes from Maurice B. Line’s 1980 comparative study for UNESCO’s General Information Programme, National Interlending Systems: A Comparative Study of Existing Systems and Possible Models. This work, still cited in current research on document delivery, identified four broad ways a country could organize its interlending. They sit on a spectrum from fully centralized to fully scattered.

Centralized collection 🔗

In this model, one central library or institution takes responsibility for collecting and supplying documents to the entire country. It maintains a deep, well-organized repository specifically built to satisfy requests. The British Library Document Supply Centre is the classic example of this approach. Because everything routes through one point, the system is efficient to manage, easy to monitor, and highly reliable for users.

Concentration on a few libraries 🔗

Here the supply role is shared among a small number of large libraries rather than one. No single institution carries the whole burden, but the responsibility still sits with a handful of well-resourced players. This spreads risk slightly and reduces the strain on any one collection, while keeping the system organized enough to remain dependable.

Planned decentralization 🔗

This model deliberately spreads supply responsibility across many libraries, but in a coordinated way. Specific libraries are assigned particular subjects or document categories, and they cooperate under an agreed structure. The goal is to create many access points without losing organization. Several national systems built around libraries holding special subject responsibilities follow this logic, where a national library works alongside designated subject specialists.

Unplanned decentralization 🔗

At the far end of the spectrum, supply happens wherever a document happens to exist, with no central planning at all. Libraries simply request from whoever holds the item. This model is flexible and adapts quickly to local needs, but it tends toward confusion. Without coordination, users may not know which library to approach, requests get duplicated, and effort is wasted.

How IFLA expanded the models in 1984 🔗

By 1984, the International Federation of Library Associations and Institutions (IFLA) had built on Line’s original framework, adding elements that addressed gaps in the earlier thinking. Two additions stand out.

Shared services 🔗

Instead of every library managing its own resources alone, multiple institutions pool materials, equipment, and infrastructure. They jointly fund and operate parts of the delivery system. This cooperative arrangement cuts costs and widens the range of documents available to any single member, which is especially valuable for smaller libraries that could never build comprehensive collections on their own.

Regional systems 🔗

The regional model organizes supply around geographic zones. A regional hub serves the libraries within its area, balancing the efficiency of centralization with the accessibility of local presence. Users benefit from specialized regional collections, and the load is distributed away from a single national center. The trade-off is that users in remote areas served by a thinly resourced region may still face delays. In a large and unevenly developed country, regional structures are an attractive middle path.

Clement’s 1988 composite models 🔗

In 1988, Clement moved the discussion toward hybrid thinking, recognizing that few real systems are purely one thing or another. The composite models allowed libraries to mix approaches depending on their size, resources, and the needs of their users. This was also the period when the IFLA Model Handbook for Interlending and Copying was published, reflecting a broader push to standardize practice. Clement’s framework rests on three components.

Networks 🔗

The network component connects libraries through technology so that catalogues, databases, and increasingly digital copies of documents can be shared quickly. A network reduces dependence on physically shipping materials and creates a responsive, interconnected web of supply. The weakness, as the IFLA analysis points out, is that a network relying on libraries with subject responsibilities can become fragile once charges are introduced, because the willingness to participate depends on funding arrangements being settled first.

Centralized back-up libraries 🔗

Even in a networked, distributed system, there is value in having a strong central library that acts as the supplier of last resort. When no network member holds a requested item, the back-up library steps in. This combines the reliability of centralization with the flexibility of a network, ensuring that hard-to-find requests do not simply fail.

Full decentralization 🔗

The third component is complete distribution, where supply is handled entirely across participating libraries with no single dominant supplier. It maximizes the number of access points and encourages local ownership, but carries the same coordination risks that affect any decentralized arrangement.

Centralized versus decentralized: weighing the trade-offs 🔗

With the models laid out, the practical comparison becomes clearer. Each approach buys certain strengths at the cost of certain weaknesses.

Strengths and limits of centralized systems 🔗

Advantages: A centralized system is efficient, easy to manage, and reliable. Resources are concentrated, so tracking holdings and processing requests is straightforward, and standards stay consistent. It is also simpler to fund and to charge for, since transactions flow through one accountable body.

Limitations: A single point of supply is also a single point of failure. If the central library is overloaded, underfunded, or physically distant from most users, delays follow. It can be expensive to build and maintain a collection deep enough to serve a whole country, and users far from the center may feel poorly served.

Strengths and limits of decentralized systems 🔗

Advantages: Decentralized systems multiply access points and bring resources closer to users. They are resilient, because the failure of one library does not collapse the whole system, and they encourage local participation and specialization. They also distribute cost across many institutions rather than concentrating it.

Limitations: Without strong coordination, decentralization breeds duplication and confusion. Users may struggle to locate the right supplier, response times vary by library, and quality control becomes harder. These are precisely the problems that planned decentralization and composite models try to fix.

How these models appear in India 🔗

India’s document delivery landscape is essentially a living blend of these models. The country runs no single dominant national supplier; instead, it relies on a planned, networked structure. INFLIBNET delivers documents through its JCCC service, having designated a set of well-stocked university libraries as interlibrary loan centres that fulfil requests for articles not held locally, a clear case of concentration on a few libraries within a planned framework. DELNET operates a resource-sharing network across thousands of member institutions, maintaining union catalogues that members search before sending requests. For social sciences, NASSDOC supplies documents from its own collection and, when an item is unavailable, sources it through interlibrary arrangements with partners such as DELNET. The result is a composite system with networks, designated supply points, and back-up routing, very much in the spirit of Clement’s 1988 thinking.

Understanding which model a system follows is more than an academic exercise. It explains why a request through one network is fast while another is slow, why some services can charge fees and others cannot, and how a country with vast geographic and resource disparities can still give a college in a small town access to materials held in a metropolitan research library.

What do you think? If you were designing a document delivery system for a country with highly uneven resources, would you lean toward a strong central library or a planned decentralized network, and what trade-offs would weigh most heavily in your decision?

References

1. https://www.sciencedirect.com/topics/social-sciences/interlibrary-loan
2. https://archive.ifla.org/IV/ifla61/61-gouu.htm
3. https://journals.stmjournals.com/joals/article=2024/view=169866/
4. https://www.inflibnet.ac.in/services/dds.php
5. http://delnet.in/delivery-services.html
6. https://icssr.org/nassdoc-document-delivery-inter-library-loan-reprography-service