View source for Preference-based planning

{{short description|Form of automated planning and scheduling}}
In [[Artificial intelligence|artificial intelligence]], '''preference-based planning''' is a form of [[Automated planning and scheduling|automated planning and scheduling]] which focuses on producing plans that additionally satisfy as many user-specified [[Preference|preference]]s as possible. In many problem domains, a task can be accomplished by various sequences of actions (also known as plans). These plans can vary in quality: there can be many ways to solve a problem but one generally prefers a way that is, e.g., cost-effective, quick and safe.

Preference-based planners take these preferences into account when producing a plan for a given problem. Examples of preference-based planning software include ''PPLAN''<ref name="pplan">[http://www.cs.toronto.edu/~sheila/PPLAN/ PPLAN], Bienvenu et al.</ref> and ''HTNPlan-P''<ref name="htnplanp">[http://www.cs.toronto.edu/kr/publications/1242050782_Sohrabi-IJCAI09.pdf HTN Planning with Preferences], Sohrabi et al.</ref> (preference-based [[Hierarchical task network|HTN planning]]).

== Overview ==

Preferences can be regarded as soft constraints on a plan. The quality of a plan increases when more preferences are satisfied but it may not be possible to satisfy all preferences in a single plan. This differs from hard constraints which must be satisfied in all plans produced by the planning software. These hard constraints are part of the [[Philosophy:Domain knowledge|domain knowledge]] while the soft constraints (or preferences) are separately specified by the user. This allows the same domain knowledge to be reused for various users who may have different preferences.

The use of preferences may also increase the length of a plan in order to satisfy more preferences. For example, when planning a journey from home to school, the user may prefer to buy a cup of coffee along the way. The planning software could now plan to visit Starbucks first and then continue to school.<ref name="starbucks">[http://www.cs.nmsu.edu/~tson/papers/tplp-plan-prefs.pdf Planning with Preferences using Logic Programming], Son and Pontelli</ref> This increases the length of the plan but the user's preference is satisfied.

== Planning Domain Definition Language ==

The [[Planning Domain Definition Language]] (as of version 3.0<ref name="pddl3">[https://dx.doi.org/10.1016/j.artint.2008.10.012 Deterministic planning in the fifth international planning competition: PDDL3 and experimental evaluation of the planners], Gerevini et al.</ref>) supports the specification of preferences through <code>preference</code> statements. For example, the statement

:<code>(preference (always (clean room1)))</code>

indicates that the user prefers that <code>room1</code> should be clean at each state of the plan. In other words, the planner should not schedule an action that causes <code>room1</code> to become dirty. As this example shows, a preference is evaluated with regard to all states of a plan (if [[Philosophy:Semantics|semantically]] required).

In addition to <code>always</code>, other constructs based on linear temporal logic are also supported, such as <code>sometime</code> (at least once during the plan), <code>sometime-after</code> (to be planned after a particular state) and <code>at-most-once</code> (the preference holds during at most one sequence of states in the plan).

=== Plan quality ===

In addition to determining whether a preference is satisfied, we also need to compute the quality of a plan based on how many preferences are satisfied. For this purpose, PDDL 3.0 includes an expression called <code>is-violated <name></code> which is equal to "the number of distinct preferences with the given name that are not satisfied in the plan".<ref name="pddl3"/> For a plan, a value can now be computed using a metric function, which is specified with <code>:metric</code>:

:<code>(:metric minimize (+ (* 5 (is-violated pref1)) (* 7 (is-violated pref2))))</code>

This example metric function specifies that the calculated value of the plan should be minimized (i.e., a plan with value ''v1'' and a plan with value ''v2'' such that ''v1'' < ''v2'', the former plan is strictly preferred). The value of a plan is computed by the given function, which is expressed in [[Polish notation]]. In this case, violation of the second preference, <code>pref2</code>, has been given a greater penalty than the first preference, <code>pref1</code>.

== Constraints satisfaction problem ==

In the area of [[Constraint satisfaction problem|constraint satisfaction problem]]s, flexible variants exist that deal with soft constraints in a similar way to preferences in preference-based planning.

==References==
{{Reflist}}

[[Category:Automated planning and scheduling]]
[[Category:Strategic management]]

{{Sourceattribution|Preference-based planning}}