Transitive binary relations  

 
indicates that the column's property is always true the row's term (at the very left), while ✗ indicates that the property is not guaranteed in general (it might, or might not, hold). For example, that every equivalence relation is symmetric, but not necessarily antisymmetric, is indicated by in the "Symmetric" column and ✗ in the "Antisymmetric" column, respectively. All definitions tacitly require the homogeneous relation be transitive: for all if and then 
In mathematics, especially order theory, a partial order on a set is an arrangement such that, for certain pairs of elements, one precedes the other. The word partial is used to indicate that not every pair of elements needs to be comparable; that is, there may be pairs for which neither element precedes the other. Partial orders thus generalize total orders, in which every pair is comparable.
Formally, a partial order is a homogeneous binary relation that is reflexive, antisymmetric, and transitive. A partially ordered set (poset for short) is an ordered pair consisting of a set (called the ground set of ) and a partial order on . When the meaning is clear from context and there is no ambiguity about the partial order, the set itself is sometimes called a poset.
The term partial order usually refers to the reflexive partial order relations, referred to in this article as nonstrict partial orders. However some authors use the term for the other common type of partial order relations, the irreflexive partial order relations, also called strict partial orders. Strict and nonstrict partial orders can be put into a onetoone correspondence, so for every strict partial order there is a unique corresponding nonstrict partial order, and vice versa.
A reflexive, weak,^{[1]} or nonstrict partial order,^{[2]} commonly referred to simply as a partial order, is a homogeneous relation ≤ on a set that is reflexive, antisymmetric, and transitive. That is, for all it must satisfy:
A nonstrict partial order is also known as an antisymmetric preorder.
An irreflexive, strong,^{[1]} or strict partial order is a homogeneous relation < on a set that is irreflexive, asymmetric, and transitive; that is, it satisfies the following conditions for all
Irreflexivity and transitivity together imply asymmetry. Also, asymmetry implies irreflexivity. In other words, a transitive relation is asymmetric if and only if it is irreflexive.^{[3]} So the definition is the same if it omits either irreflexivity or asymmetry (but not both).
A strict partial order is also known as an asymmetric strict preorder.
Strict and nonstrict partial orders on a set are closely related. A nonstrict partial order may be converted to a strict partial order by removing all relationships of the form that is, the strict partial order is the set where is the identity relation on and denotes set subtraction. Conversely, a strict partial order < on may be converted to a nonstrict partial order by adjoining all relationships of that form; that is, is a nonstrict partial order. Thus, if is a nonstrict partial order, then the corresponding strict partial order < is the irreflexive kernel given by Conversely, if < is a strict partial order, then the corresponding nonstrict partial order is the reflexive closure given by:
The dual (or opposite) of a partial order relation is defined by letting be the converse relation of , i.e. if and only if . The dual of a nonstrict partial order is a nonstrict partial order,^{[4]} and the dual of a strict partial order is a strict partial order. The dual of a dual of a relation is the original relation.
Given a set and a partial order relation, typically the nonstrict partial order , we may uniquely extend our notation to define four partial order relations and , where is a nonstrict partial order relation on , is the associated strict partial order relation on (the irreflexive kernel of ), is the dual of , and is the dual of . Strictly speaking, the term partially ordered set refers to a set with all of these relations defined appropriately. But practically, one need only consider a single relation, or , or, in rare instances, the nonstrict and strict relations together, .^{[5]}
The term ordered set is sometimes used as a shorthand for partially ordered set, as long as it is clear from the context that no other kind of order is meant. In particular, totally ordered sets can also be referred to as "ordered sets", especially in areas where these structures are more common than posets. Some authors use different symbols than such as ^{[6]} or ^{[7]} to distinguish partial orders from total orders.
When referring to partial orders, should not be taken as the complement of . The relation is the converse of the irreflexive kernel of , which is always a subset of the complement of , but is equal to the complement of if, and only if, is a total order.^{[a]}
Another way of defining a partial order, found in computer science, is via a notion of comparison. Specifically, given as defined previously, it can be observed that two elements x and y may stand in any of four mutually exclusive relationships to each other: either x < y, or x = y, or x > y, or x and y are incomparable. This can be represented by a function that returns one of four codes when given two elements.^{[8]}^{[9]} This definition is equivalent to a partial order on a setoid, where equality is taken to be a defined equivalence relation rather than set equality.^{[10]}
Wallis defines a more general notion of a partial order relation as any homogeneous relation that is transitive and antisymmetric. This includes both reflexive and irreflexive partial orders as subtypes.^{[1]}
A finite poset can be visualized through its Hasse diagram.^{[11]} Specifically, taking a strict partial order relation , a directed acyclic graph (DAG) may be constructed by taking each element of to be a node and each element of to be an edge. The transitive reduction of this DAG^{[b]} is then the Hasse diagram. Similarly this process can be reversed to construct strict partial orders from certain DAGs. In contrast, the graph associated to a nonstrict partial order has selfloops at every node and therefore is not a DAG; when a nonstrict order is said to be depicted by a Hasse diagram, actually the corresponding strict order is shown.
Standard examples of posets arising in mathematics include:
One familiar example of a partially ordered set is a collection of people ordered by genealogical descendancy. Some pairs of people bear the descendantancestor relationship, but other pairs of people are incomparable, with neither being a descendant of the other.
In order of increasing strength, i.e., decreasing sets of pairs, three of the possible partial orders on the Cartesian product of two partially ordered sets are (see Fig. 4):
All three can similarly be defined for the Cartesian product of more than two sets.
Applied to ordered vector spaces over the same field, the result is in each case also an ordered vector space.
See also orders on the Cartesian product of totally ordered sets.
Another way to combine two (disjoint) posets is the ordinal sum^{[12]} (or linear sum),^{[13]} Z = X ⊕ Y, defined on the union of the underlying sets X and Y by the order a ≤_{Z} b if and only if:
If two posets are wellordered, then so is their ordinal sum.^{[14]}
Seriesparallel partial orders are formed from the ordinal sum operation (in this context called series composition) and another operation called parallel composition. Parallel composition is the disjoint union of two partially ordered sets, with no order relation between elements of one set and elements of the other set.
The examples use the poset consisting of the set of all subsets of a threeelement set ordered by set inclusion (see Fig. 1).
There are several notions of "greatest" and "least" element in a poset notably:
As another example, consider the positive integers, ordered by divisibility: 1 is a least element, as it divides all other elements; on the other hand this poset does not have a greatest element. This partially ordered set does not even have any maximal elements, since any g divides for instance 2g, which is distinct from it, so g is not maximal. If the number 1 is excluded, while keeping divisibility as ordering on the elements greater than 1, then the resulting poset does not have a least element, but any prime number is a minimal element for it. In this poset, 60 is an upper bound (though not a least upper bound) of the subset which does not have any lower bound (since 1 is not in the poset); on the other hand 2 is a lower bound of the subset of powers of 2, which does not have any upper bound. If the number 0 is included, this will be the greatest element, since this is a multiple of every integer (see Fig. 6).
Given two partially ordered sets (S, ≤) and (T, ≼), a function is called orderpreserving, or monotone, or isotone, if for all implies f(x) ≼ f(y). If (U, ≲) is also a partially ordered set, and both and are orderpreserving, their composition is orderpreserving, too. A function is called orderreflecting if for all f(x) ≼ f(y) implies If f is both orderpreserving and orderreflecting, then it is called an orderembedding of (S, ≤) into (T, ≼). In the latter case, f is necessarily injective, since implies and in turn according to the antisymmetry of If an orderembedding between two posets S and T exists, one says that S can be embedded into T. If an orderembedding is bijective, it is called an order isomorphism, and the partial orders (S, ≤) and (T, ≼) are said to be isomorphic. Isomorphic orders have structurally similar Hasse diagrams (see Fig. 7a). It can be shown that if orderpreserving maps and exist such that and yields the identity function on S and T, respectively, then S and T are orderisomorphic.^{[15]}
For example, a mapping from the set of natural numbers (ordered by divisibility) to the power set of natural numbers (ordered by set inclusion) can be defined by taking each number to the set of its prime divisors. It is orderpreserving: if x divides y, then each prime divisor of x is also a prime divisor of y. However, it is neither injective (since it maps both 12 and 6 to ) nor orderreflecting (since 12 does not divide 6). Taking instead each number to the set of its prime power divisors defines a map that is orderpreserving, orderreflecting, and hence an orderembedding. It is not an orderisomorphism (since it, for instance, does not map any number to the set ), but it can be made one by restricting its codomain to Fig. 7b shows a subset of and its isomorphic image under g. The construction of such an orderisomorphism into a power set can be generalized to a wide class of partial orders, called distributive lattices; see Birkhoff's representation theorem.
Sequence A001035 in OEIS gives the number of partial orders on a set of n labeled elements:
Elements  Any  Transitive  Reflexive  Symmetric  Preorder  Partial order  Total preorder  Total order  Equivalence relation 

0  1  1  1  1  1  1  1  1  1 
1  2  2  1  2  1  1  1  1  1 
2  16  13  4  8  4  3  3  2  2 
3  512  171  64  64  29  19  13  6  5 
4  65,536  3,994  4,096  1,024  355  219  75  24  15 
n  2^{n2}  2^{n(n−1)}  2^{n(n+1)/2}  ∑^{n} _{k=0} k!S(n, k) 
n!  ∑^{n} _{k=0} S(n, k)  
OEIS  A002416  A006905  A053763  A006125  A000798  A001035  A000670  A000142  A000110 
Note that S(n, k) refers to Stirling numbers of the second kind.
The number of strict partial orders is the same as that of partial orders.
If the count is made only up to isomorphism, the sequence 1, 1, 2, 5, 16, 63, 318, ... (sequence A000112 in the OEIS) is obtained.
A poset is called a subposet of another poset provided that is a subset of and is a subset of . The latter condition is equivalent to the requirement that for any and in (and thus also in ), if then .
If is a subposet of and furthermore, for all and in , whenever we also have , then we call the subposet of induced by , and write .
A partial order on a set is called an extension of another partial order on provided that for all elements whenever it is also the case that A linear extension is an extension that is also a linear (that is, total) order. As a classic example, the lexicographic order of totally ordered sets is a linear extension of their product order. Every partial order can be extended to a total order (orderextension principle).^{[16]}
In computer science, algorithms for finding linear extensions of partial orders (represented as the reachability orders of directed acyclic graphs) are called topological sorting.
Every poset (and every preordered set) may be considered as a category where, for objects and there is at most one morphism from to More explicitly, let hom(x, y) = {(x, y)} if x ≤ y (and otherwise the empty set) and Such categories are sometimes called posetal.
Posets are equivalent to one another if and only if they are isomorphic. In a poset, the smallest element, if it exists, is an initial object, and the largest element, if it exists, is a terminal object. Also, every preordered set is equivalent to a poset. Finally, every subcategory of a poset is isomorphismclosed.
If is a partially ordered set that has also been given the structure of a topological space, then it is customary to assume that is a closed subset of the topological product space Under this assumption partial order relations are well behaved at limits in the sense that if and and for all then ^{[17]}
A convex set in a poset P is a subset I of P with the property that, for any x and y in I and any z in P, if x ≤ z ≤ y, then z is also in I. This definition generalizes the definition of intervals of real numbers. When there is possible confusion with convex sets of geometry, one uses orderconvex instead of "convex".
A convex sublattice of a lattice L is a sublattice of L that is also a convex set of L. Every nonempty convex sublattice can be uniquely represented as the intersection of a filter and an ideal of L.
An interval in a poset P is a subset that can be defined with interval notation:
Whenever a ≤ b does not hold, all these intervals are empty. Every interval is a convex set, but the converse does not hold; for example, in the poset of divisors of 120, ordered by divisibility (see Fig. 7b), the set {1, 2, 4, 5, 8} is convex, but not an interval.
An interval I is bounded if there exist elements such that I ⊆ [a, b]. Every interval that can be represented in interval notation is obviously bounded, but the converse is not true. For example, let P = (0, 1) ∪ (1, 2) ∪ (2, 3) as a subposet of the real numbers. The subset (1, 2) is a bounded interval, but it has no infimum or supremum in P, so it cannot be written in interval notation using elements of P.
A poset is called locally finite if every bounded interval is finite. For example, the integers are locally finite under their natural ordering. The lexicographical order on the cartesian product is not locally finite, since (1, 2) ≤ (1, 3) ≤ (1, 4) ≤ (1, 5) ≤ ... ≤ (2, 1). Using the interval notation, the property "a is covered by b" can be rephrased equivalently as
This concept of an interval in a partial order should not be confused with the particular class of partial orders known as the interval orders.
So we can think of every partial order as really being a pair, consisting of a weak partial order and an associated strict one.
compare_elements(x, y): Compare x and y in the poset. If x < y, return −1. If x = y, return 0. If x > y, return 1. If x and y are not comparable, return None.
A comparison between two elements s, t in S returns one of three distinct values, namely s≤t, s>t or st.
A partially ordered set is conveniently represented by a Hasse diagram...
Key concepts  

Results  
Properties & Types (list) 

Constructions  
Topology & Orders  
Related 