Combinatorics: High-Frequency Difficulties Explained | 组合数学高频难点解析

📚 Combinatorics: High-Frequency Difficulties Explained | 组合数学高频难点解析

Combinatorics often does not ask you to perform a single mechanical calculation. It tests whether you can translate a wordy situation into the correct counting model. The most common errors are subtle: choosing the wrong operation, forgetting identical objects, and over-splitting or double-counting cases. This article breaks down the high-frequency difficulties that keep appearing in exams.

组合数学通常不是只让你做一次机械计算。它考查的是能否将文字情境转化为正确的计数模型。最常见的错误非常隐蔽：选错运算、忘记相同物品、重复计数或分类过细。本文将系统拆解考试中反复出现的高频难点。

---

1. The first step: does order matter? | 第一步：顺序是否重要？

The most important decision in every combinatorics problem is whether the arrangement order matters. If changing the order creates a different outcome, you are counting permutations. If the order is irrelevant and you are only choosing a group, you are counting combinations.

每一道组合题中最关键的判断就是“顺序是否重要”。如果交换顺序会形成不同的结果，你就在计算排列数；如果顺序无关紧要，你只是选出一个组，那么你就在计算组合数。

For example, choosing 3 students from 5 to form a committee gives C(5,3) = 10, because the committee has no ranks. But choosing 3 students to be president, secretary and treasurer gives P(5,3) = 60, because the same three names can be arranged in different positions.

例如，从 5 名学生中选 3 人组成委员会，结果是 C(5,3) = 10，因为委员会内部没有职务区别。但如果从 5 名学生中选出 3 人分别担任主席、书记和财务，结果是 P(5,3) = 60，因为同样的三个人可以分配到不同职位。

A useful check is to ask: would swapping two selected objects ever create a new case? If yes, use a permutation. If not, use a combination.

一个有效的检验方法是：交换两个被选出的对象是否会产生新情况？如果会，就用排列；如果不会，就用组合。

---

2. Using nPr and nCr correctly | 正确使用排列数与组合数

For n distinct objects and no repetition, the number of ordered arrangements of r objects is P(n,r), while the number of unordered selections of r objects is C(n,r). The definitions must be handled carefully.

当 n 个物体互不相同且不允许重复时，从 n 个中取出 r 个的有序排列数为 P(n,r)，而无序选择数为 C(n,r)。这两个定义必须准确掌握。

P(n,r) = n! / (n – r)!, C(n,r) = n! / [r!(n – r)!]

In older notation these are written nPr and nCr. When using a calculator, be sure that the value entered is the smaller number r, not the difference n – r.

在旧记号中，它们写作 nPr 和 nCr。使用计算器时，要确认输入的是较小的数 r，而不是差值 n – r。

One important identity is C(n,r) = C(n,n-r). For example, C(20,17) = C(20,3) = 1140. This symmetry is not true for permutations, since P(n,n-r) is different from P(n,r).

一个重要恒等式是 C(n,r) = C(n,n-r)。例如，C(20,17) = C(20,3) = 1140。这种对称性对排列不成立，因为 P(n,n-r) 与 P(n,r) 不相等。

---

3. Arrangements with identical objects | 含有相同元素的排列

When objects are not all distinct, a direct factorial overcounts. You must divide by the factorial of the number of identical objects in each repeated group.

当物体不完全相同时，直接用阶乘会重复计数。你必须除以每一种重复物体数量的阶乘。

For the word BANANA, there are 6 letters in total: B appears 1 time, A appears 3 times, and N appears 2 times. The number of distinct arrangements is 6! / (3! × 2!) = 60.

以单词 BANANA 为例，共有 6 个字母：B 出现 1 次，A 出现 3 次，N 出现 2 次。不同的排列数为 6! / (3! × 2!) = 60。

The reason is that swapping two identical letters does not produce a new arrangement. Dividing by the duplicate factorials removes all these false repetitions.

原因是交换两个相同字母不会产生新的排列。除以重复部分的阶乘，就是要去掉所有这类多余计数。

This idea also appears in arrangements of coloured balls and routes on a grid that must use a fixed number of right and up moves.

这一思想也出现在彩色球的排列中，以及固定数量的“向右走”和“向上走”所构成的路径计数问题中。

---

4. Grouping and distributing objects | 分组与分配问题

Grouping problems are tricky because students forget whether the groups are labelled or unlabelled. If the groups have names, you multiply by the permutations of the group labels. If the groups are not named, you must divide by the factorial of the repeated group sizes.

分组问题的难点在于学生经常忘记分组到底有没有标签。如果各组有名称，就需要乘以组标签的排列数；如果各组没有名称，则必须除以相等组数的阶乘。

For example, dividing 6 different books into two labelled boxes, with 3 books in each box, gives C(6,3) = 20. But dividing the same books into two unlabelled stacks of 3 gives C(6,3) / 2! = 10.

例如，把 6 本不同的书放入两个有标签的盒子，每个盒子 3 本，共有 C(6,3) = 20 种方法。但如果只是分成两摞，每摞 3 本，并且两摞没有标签，则为 C(6,3) / 2! = 10 种。

A common formula for splitting n distinct items into k unlabelled groups of equal size m is n! / [(m!)^k × k!].

把 n 个不同物品分成 k 个相同大小的无标签组，每组 m 个，常用公式为 n! / [(m!)^k × k!]。

Example: 6 items into 3 unlabelled groups of 2 → 6! / [(2!)^3 × 3!] = 15

When the groups are labelled, the denominator does not contain k!, so the same division would give 90.

如果各组有标签，分母中就不包含 k!，所以同样的分组方法数为 90。

---

5. Adjacency restrictions | 相邻限制问题

When certain items must be together, the standard method is the block method. Treat the forced group as one single object, arrange all objects, and then arrange the items inside the block.

当某些物品必须相邻时，标准方法是“捆绑法”。先把必须相邻的一组物品看作一个整体对象，再对所有对象进行排列，最后排列整体内部的物品。

For example, arrange 6 people in a row so that two specific friends must sit next to each other. The two friends form one block, so there are 5 objects to arrange: 5! ways. Inside the block, the two friends can switch places: 2! ways. The total is 5! × 2! = 240.

例如，6 个人排成一排，其中两个好朋友必须相邻。把这两个人看成一个整体，于是有 5 个对象需要排列：5! 种。在这个整体内部，两人可以互换位置：2! 种。总数为 5! × 2! = 240。

If the two friends must not sit next to each other, subtract the adjacent case from the total: 6! – 5! × 2! = 720 – 240 = 480.

如果这两个人不能相邻，则从总数中减去相邻的情况：6! – 5! × 2! = 720 – 240 = 480。

This complement method is often faster than counting separated positions directly.

这种“补集法”通常比直接枚举不相邻位置更快。

---

6. Circular arrangements | 环形排列

Circular arrangement questions require a different formula because rotating the whole circle does not change the arrangement. For n distinct objects placed around a circle, the number of arrangements is (n-1)!.

环形排列需要使用不同公式，因为旋转整个圆桌不会改变排列。将 n 个不同物体放在一个圆上，排列数为 (n-1)!。

For example, 5 people around a circular table can be arranged in 4! = 24 ways. If the seats are numbered or labelled, no rotation is equivalent, so the answer becomes 5! = 120.

例如，5 个人围圆桌而坐，排列数为 4! = 24。如果座位有编号或有位置标签，旋转不再等价，答案就是 5! = 120。

If the circular object can be flipped over, such as a necklace or a bracelet, mirror images become identical. The count is then divided by 2, giving (n-1)! / 2.

如果圆形对象可以翻转，例如项链或手链，则镜像也被视为相同，因此数量要除以 2，得到 (n-1)! / 2。

Thus 5 different beads on a necklace can be arranged in 4! / 2 = 12 ways.

因此，5 颗不同珠子串成一条项链共有 4! / 2 = 12 种串法。

---

7. The “at least one” trap | “至少一个”的陷阱

Phrases such as “at least one” often hide a complement. The safest approach is to calculate the total number of cases and subtract the undesirable cases with zero selected.

“至少一个”这类表述往往隐藏着补集技巧。最安全的方法是计算总情况数，再减去“一个都不选”的不可接受情况。

For 5 distinct fruits, the number of ways to choose at least one is 2⁵ – 1 = 31. The term 2⁵ counts every fruit as either chosen or not chosen, and the minus 1 removes the empty set.

对于 5 种不同水果，至少选一种的方法数为 2⁵ – 1 = 31。2⁵ 表示每种水果都被“选”或“不选”，减去的 1 是去掉什么都不选的情况。

A typical committee problem: choose 4 students from 5 boys and 4 girls, with at least one girl. The total is C(9,4) = 126, and the no-girl case is C(5,4) = 5. The answer is 121.

一个典型委员会问题是：从 5 名男生和 4 名女生中选 4 人，且至少有 1 名女生。总数为 C(9,4) = 126，没有女生的情况为 C(5,4) = 5，因此答案是 121。

When using “at least one”, avoid splitting into too many cases if a complement is available.

在处理“至少一个”时，只要能使用补集，就不要把事情拆成过多小类。

---

8. Selections with repetition | 允许重复选择的组合

Some problems allow the same type of object to be chosen more than once. The number of ways to choose r items from n types, when repetition is allowed and order does not matter, is C(n + r – 1, r).

有些问题允许同一类对象被多次选择。当允许重复且顺序不重要时，从 n 类物品中选择 r 个物品的方法数为 C(n + r – 1, r)。

This is also the number of non-negative integer solutions to x₁ + x₂ + … + xₙ = r.

这也可以理解为方程 x₁ + x₂ + … + xₙ = r 的非负整数解个数。

For example, distributing 10 identical sweets among 4 children, where a child may receive zero sweets, gives C(4 + 10 – 1, 10) = C(13, 10) = 286.

例如，把 10 颗完全相同的糖果分给 4 个孩子，允许某个孩子分到 0 颗，则方法数为 C(4 + 10 – 1, 10) = C(13, 10) = 286。

If each child must receive at least one sweet, first give each child 1 sweet and then distribute the remaining 6 sweets freely: C(4 + 6 – 1, 6) = C(9,6) = 84.

如果每个孩子至少分到 1 颗，则先给每个孩子 1 颗，再把剩下的 6 颗自由分配：C(4 + 6 – 1, 6) = C(9,6) = 84。

This method is often called stars and bars, though the formula itself is usually enough in an exam.

这种方法常被称为“隔板法”或“星与条法”，但考试时掌握公式本身通常就足够了。

---

9. Inclusion-exclusion for overlapping categories | 容斥原理处理重叠类别

When two groups overlap, simply adding their sizes double-counts the intersection. The inclusion-exclusion principle corrects this error.

当两个集合有重合时，直接把两个集合大小相加会把交集重复计算。容斥原理正是用来修正这一错误。

|A ∪ B| = |A| + |B| – |A ∩ B|

For example, in a class of 30 students, 18 study mathematics, 15 study physics, and 10 study both. The number studying at least one subject is 18 + 15 – 10 = 23. The number studying neither is 30 – 23 = 7.

例如，某班有 30 名学生，18 人学数学，15 人学物理，10 人两科都学。至少学一科的人数为 18 + 15 – 10 = 23。两科都不学的人数为 30 – 23 = 7。

For three sets, the formula extends to adding all single sets, subtracting all pair intersections, and adding the triple intersection back once.

对于三个集合，公式扩展为：加上所有单个集合，减去所有两两交集，再加上三者交集。

|A ∪ B ∪ C| = |A| + |B| + |C| – |A∩B| – |A∩C| – |B∩C| + |A∩B∩C|

This pattern is especially useful when counting numbers divisible by 2, 3 or 5, or when counting students taking multiple subjects.

这一模式在统计能被 2、3、5 整除的数，或统计选了多门科目的学生时非常有用。

---

10. Binomial theorem and Pascal identity | 二项式定理与帕斯卡恒等式

Combinatorics is closely connected to algebra through the binomial theorem. The combination number C(n,k) appears as the coefficient in the expansion of a binomial.

组合数学与代数通过二项式定理紧密相连。组合数 C(n,k) 出现在二项式展开的系数中。

(a + b)ⁿ = Σ C(n,k)aⁿ⁻ᵏbᵏ, for k = 0 to n

For example, (1 + x)⁴ = C(4,0) + C(4,1)x + C(4,2)x² + C(4,3)x³ + C(4,4)x⁴ = 1 + 4x + 6x

Published by TutorHao | Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）