Back to the Research Activities

The Office Space Allocation Problem - Overview

Description

The class of Space Allocation or Capacity Allocation problems are those in which the amount of space (area or volume) or capacity that is available has to be distributed among a set of items, satisfying specific requirements and constraints. Examples of this class of problems are: bin packing problem, knapsack problem, space planning and others.

Here, Office Space Allocation refers to the distribution of the available areas of office space among a number of entities with different sizes so as to ensure the optimal space utilisation and the satisfaction of additional requirements and/or constraints. In this generic case, an important condition exists: the areas of space that can be used and the space required by the entities are not subject to modification. Then the quality of a solution or allocation is measured in terms of the following five aspects (not necessarily in this order of importance):

• the number of entities than have been allocated.

• space utilisation, i.e the amount of space that is wasted (space not used) and the amount of space that is overused (entities with less space allocated than needed).

• satisfaction of any additional requirements.

• satisfaction or no violation of constraints.

The ideal solution in the space allocation problem is one where all the entities are allocated, no space is wasted or overused and every additional requirements and constraints have been satisfied. However, not always this ideal optimal solution is achievable. In a more realistic scenario, the optimal solution would be one where all objects are scheduled and the space utilisation is the best possible, i.e. the amount of space wasted and overused has been reduced to the minimum and the additional requirements and constraints have been all satisfied.

Some examples of constraints (specific restrictions hat should or must be fulfilled) are listed below, but different constraints may exist in different scenarios. Constraints can be classified as hard or soft. Hard constraints are those that cannot be violated while soft constraints are those that can be broken but penalised. To minimise the penalties in a solution for an Office Space Allocation problem, no hard constraints should be violated and as many as possible soft constraints should be satisfied.

• Wastage. Restricts the amount of space that can be not used in each area.
• Overuse. Refers to the amount of space that can be overused, i.e. the difference between the space available and the space needed for the allocated entities.
• Unallocated. Refers to those entities that are not allocated in the solution.
• Sharing. Restricts sharing a common area between two or more entities, i.e. refers to entities that must not be allocated in the same office space.
• Be located in. Restricts the allocation of a particular entity to the indicated preferred area.
• Be adjacent to. Refers to the situation in which some entities have to be allocated in adjacent areas.
• Be away of. Specifies that some entities have to be allocated away of a certain entities or areas.
• Be together with. Refers to the situation in which some entities have to be allocated in the same location.
• Be grouped with. Refers to the situation in which some entities have to be grouped, i.e. the entities will be in nearby areas.
• Not overused. Specifies that certain areas cannot be overused.
• Disturbance. Restrict the amount of changes when reorganising an existing solution.

Most of the real instances of the Office Space Allocation problem can be classified as one of the following types:

Reorganisation of the existing allocation. Refers to the rearrangement of the current space distribution among the entities and it is performed when either it is required to improve the existing solution under the existing conditions or it is required to modify the allocation because the conditions (requirements, constraints, number of entities to be allocated, number of areas of space) of the problem change.

Construction of a complete solution. The construction of a complete allocation or solution refers to the generation of a new solution from scratch to distribute all the available areas of space among all the entities in the problem under the given conditions.

When reorganising an existing allocation, it may be required to minimise the amount of disruption caused, i.e. relocation of entities. This constraint exists because it may be too costly to move every entity around and this often impedes finding the optimal utilisation of space. When reorganising an allocation, the amount of disruption permitted establishes a balance between the quality of the allocation and the difficulty in achieving it.

Office Space Allocation in Academic Institutions

In the context of higher educational institutions, this problem is defined as the allocation of entities (staff, students, meeting rooms, lecture rooms, special rooms, etc.) to areas of space such as rooms, satisfying as many requirements and constrains as possible. This problem is highly constrained, has multiple objectives, varies greatly among different institutions, requires frequent modifications due to the addition or remove of entities and/or rooms and has a direct impact on the functionality of the university.

Allocating rooms in universities is a multi-stage process:

1. The centralised office or committee allocates space to faculties and assigns common areas.

2. Faculties assign areas to schools and departments.

3. Departments allocate specific rooms to entities.

During these three phases, the problem can be solved in different ways at each stage:

• Fitting all entities into a limited amount of space.

• Minimising the amount of space required to allocate a set of entities.

• Reorganising because of the addition or removal space and/or entities.

• Reorganising the current allocation due to the variation of requirements.

• Reorganising because of the modification of rooms (moving walls for example).

Related Information

• The Automated Office Space Allocation is a prototype software for assisting on the construction of allocations.
• The test data sets for this project were created from real data collected from British Universities.
• The publications, including the PhD Thesis in pdf format, related to the space allocation research project are available here.