📚 Patterns of Urbanisation Since 1945: A Mathematical Modelling Approach | 1945年以来城市化模式：数学建模方法
The global urban landscape has transformed dramatically since the end of the Second World War. While geographers and sociologists study the causes and consequences of urbanisation, mathematics provides powerful tools to quantify, model, and predict these patterns. By examining urban growth through differential equations, statistical regression, and logistic curves, we can uncover the underlying dynamics of how cities have expanded across different regions and periods. This article explores the mathematical modelling of urbanisation since 1945, linking concepts from Edexcel A-Level Mathematics to real-world demographic change.
自二战结束以来，全球城市格局发生了巨大变化。虽然地理学家和社会学家研究城市化的原因和后果，但数学提供了量化、建模和预测这些模式的强大工具。通过微分方程、统计回归和逻辑斯蒂曲线来审视城市增长，我们可以揭示不同地区和时期城市扩张的内在动力学。本文探讨1945年以来城市化的数学建模，将Edexcel A-Level数学中的概念与现实世界的人口变化联系起来。
1. Post-War Urban Explosion: Setting the Scene | 战后城市大爆炸：背景概述
In 1950, approximately 30% of the world's population lived in urban areas. By 2020, that figure had surpassed 56%, driven by industrialisation, rural-to-urban migration, and natural increase. Mathematically, such massive growth invites us to describe the changing proportion of urban dwellers over time using functions and rates of change.
1950年，全球约30%的人口居住在城市地区。到2020年，这一数字已超过56%，其驱动力包括工业化、农村向城市迁移以及自然增长。从数学角度看，如此大规模的增长促使我们使用函数和变化率来描述城市人口比例随时间的变化。
The post-1945 period is ideal for constructing piecewise models because the pace of urbanisation has not been uniform. A simple linear model fails to capture the acceleration in Asia after 1980 or the levelling-off in Western Europe. Instead, we apply exponential, logistic, and hybrid functions to fit observed data.
1945年后这段时间非常适合构建分段模型，因为城市化速度并非均匀。简单的线性模型无法捕捉1980年后亚洲的加速增长或西欧的趋于平稳。因此，我们应用指数函数、逻辑斯蒂函数和混合函数来拟合观测数据。
2. Exponential Growth as a First Approximation | 指数增长作为初步近似
opulation and k is a positive constant. The solution is an exponential function: U(t) = U₀ eᵏᵗ, with U₀ being the population at t = 0.
如果城市化速率与当前城市人口成正比，我们可以写出微分方程 dU/dt = kU，其中U表示城市人口，k是一个正常数。该方程的解为指数函数：U(t) = U₀ eᵏᵗ，其中U₀为t = 0时的人口。
For many developing countries during the 1950s and 1960s, exponential growth provided a reasonable short-term forecast. For instance, Mexico City's population grew from about 3 million in 1950 to over 9 million by 1970, corresponding to an average annual growth rate of roughly 5.5%. Using an exponential model, we can estimate the doubling time via the formula Tdouble = ln 2 / k.
对于20世纪50年代和60年代的许多发展中国家，指数增长提供了一个合理的短期预测。例如，墨西哥城的人口从1950年约300万增长到1970年超过900万，对应的年均增长率约为5.5%。利用指数模型，我们可以通过公式 Tdouble = ln 2 / k 估算倍增时间。
However, exponential growth cannot continue indefinitely. Resources, space, and economic constraints impose limits, which leads us to more refined models.
然而，指数增长不可能无限持续。资源、空间和经济约束会施加限制，这促使我们使用更精细的模型。
3. The Logistic Model and Carrying Capacity | 逻辑斯蒂模型与承载容量
1 − U/K), where r is the intrinsic growth rate and K is the carrying capacity (maximum sustainable urban population). The solution forms an S-shaped sigmoid curve, often called the urban transition curve.
经典的改进是逻辑斯蒂微分方程：dU/dt = rU(1 − U/K)，其中r为内在增长率，K为承载容量（最大可持续城市人口）。其解形成一条S形曲线，常被称为城市转型曲线。
ation that can live in cities given technology and economic structure. For example, the UK' the 19th and early 20th centuries, but after 1950 it decelerated and has hovered around 83% since the 1980s, suggesting K ≈ 0.83.
在城市化的背景下，K不一定代表绝对上限，而是指在技术和经济结构条件下，能够长期居住在城市的人口均衡比例。例如，英国的城市化比例在19世纪和20世纪初迅速增长，但在1950年后减速，自1980年代起一直徘徊在83%左右，表明K ≈ 0.83。
U₀) − rt. This allows students to use linear regression to estimate r and verify the model's validity.
为了将逻辑斯蒂曲线拟合到真实数据上，我们通常会使用变换使其线性化：ln((K−U)/U) = ln((K−U₀)/U₀) − rt。这使得学生可以使用线性回归来估算r并验证模型的有效性。
4. Differential Equations in Urbanisation Modelling | 城市化建模中的微分方程
Urbanisation models often employ coupled differential equations to capture interactions between urban and rural populations. Let Pu be urban population and Pr be rural population. A simple system might be:
dP_u/dt = a Pu + b Pr − c Pu
dP_r/dt = d Pr − e Pr
with migration and birth/death rates embedded in the parameters.
城市化模型常采用耦合微分方程来捕捉城市与农村人口之间的相互作用。设Pu为城市人口，Pr为农村人口。一个简单系统可能是上面所示，其中迁移、出生率和死亡率都包含在参数中。
These systems can be analysed using techniques from A-Level Further Mathematics, such as phase plane analysis, equilibrium points, and stability. For instance, a stable equilibrium (Pu*, Pr*) might represent a fully urbanised nation. Understanding these helps explain why some countries urbanise faster than others.
这些系统可以使用A-Level进阶数学中的技术进行分析，例如相平面分析、平衡点和稳定性。例如，一个稳定平衡点(Pu*, Pr*)可以代表一个完全城市化的国家。理解这些有助于解释为何一些国家的城市化速度快于其他国家。
5. Data Analysis: Urban Population Data Since 1945 | 数据分析：1945年以来城市人口数据
Reliable data from the United Nations World Urbanization Prospects provide annual estimates of urban population and urbanisation share for every country since 1950. These datasets allow students to practise statistical techniques such as determining the line of best fit, calculating correlation coefficients, and computing moving averages to smooth short-term fluctuations.
来自联合国《世界城市化展望》的可靠数据提供了自1950年以来每个国家的城市人口和城市化比例的年度估算。这些数据集让学生可以练习统计技术，如确定最佳拟合线、计算相关系数，以及计算移动平均以平滑短期波动。
Consider the table below showing the urban share for three regions at ten-year intervals:
| Year | Sub-Saharan Africa | East Asia | Europe |
|---|---|---|---|
| 1950 | 11.2% | 16.8% | 51.3% |
| 1980 | 22.7% | 27.0% | 66.0% |
| 2020 | 41.1% | 64.8% | 74.9% |
Using this data, we can compute average decadal growth rates. East Asia's urban share grew by approximately 2.3 percentage points per decade initially, but accelerated dramatically after the 1980s due to China's economic reforms, illustrating a non-linear trend.
利用这些数据，我们可以计算每十年的平均增长率。东亚的城市化比例最初每十年增长约2.3个百分点，但在20世纪80年代后因中国经济改革而急剧加速，显示出非线性趋势。
6. The S-Curve of Urban Transition | 城市化转型的S曲线
ulative distribution of a logistic function. The inflection point occurs where the curve switches from concave up to concave down, which is when U = K/2. At this point, the growth rate dU/dt is maximised.
城市化转型理论认为，各国沿着一条S曲线从低城市化水平走向高城市化水平。从数学上讲，这被逻辑斯蒂函数的累积分布所捕捉。拐点出现在曲线从凹向上变为凹向下的位置，即当U = K/2时。在这一点上，增长率dU/dt达到最大。
Identifying an inflection point in a country's urbanisation data helps policymakers anticipate peak pressure on infrastructure. For example, China's urbanisation share exceeded 50% around 2011, placing it near the inflection point, which corresponded with massive investment in high-speed rail and urban housing.
识别一国城市化数据中的拐点有助于政策制定者预测基础设施压力的高峰期。例如，中国的城市化比例在2011年前后超过50%，使其接近拐点，这与高铁和城市住房的大规模投资相吻合。
the year when U = K/2. Differentiation and using the chain rule are essential skills here.
我们可以通过拟合逻辑斯蒂曲线并求解U = K/2的年份来估算拐点日期。微分和链式法则是这里的基本技能。
7. Regression Techniques for Trend Fitting | 趋势拟合的回归技术
ential model U = a eᵇᵗ, we take natural logarithms: ln U = ln a + b t. For a power-law model U = c tᵈ, we use log-log transformation: ln U = ln c + d ln t.
线性回归是A-Level统计学的基石。为了对城市化建模，我们常常对变量进行变换以实现线性。对于指数模型 U = a eᵇᵗ，我们取自然对数：ln U = ln a + b t。对于幂律模型 U = c tᵈ，我们使用对数-对数变换：ln U = ln c + d ln t。
By performing regression on transformed data, we can calculate the coefficient of determination R², which indicates how well the model explains the variability. A high R² (e.g., > 0.95) for a logistic linearised form confirms the S-curve nature of urbanisation. Students should be aware of the dangers of extrapolation beyond the data range, as urban share cannot exceed 100%.
通过对变换后的数据进行回归，我们可以计算决定系数R²，它表明模型解释变异性的程度。对于逻辑斯蒂线性化形式，高的R²（例如>0.95）证实了城市化的S曲线性质。学生应意识到将数据外推超出范围的危险，因为城市化比例不可能超过100%。
8. Limitations and Stochastic Modelling | 局限性与随机建模
ferential equations (SDEs) or add noise terms to capture uncertainty: dU = rU(1−U/K) dt + σU dW, where dW is a Wiener process.
确定性模型忽略了战争、经济危机和自然灾害等随机冲击。在数学中，我们可以引入随机微分方程或添加噪声项来捕捉不确定性：dU = rU(1−U/K) dt + σU dW，其中dW是维纳过程。
While the full treatment of SDEs is beyond standard A-Level, the concept of residual analysis is accessible. After fitting a deterministic model, plotting residuals can reveal patterns indicating missing variables, such as migration policies or technological change.
虽然随机微分方程的完整处理超出了常规A-Level范围，但残差分析的概念是可行的。在拟合确定性模型后，绘制残差图可以揭示出指示遗漏变量（如移民政策或技术变革）的模式。
Furthermore, all models are simplifications. The assumption of a constant carrying capacity K may break down if technology shifts, e.g., the internet reducing the need for physical urban concentration.
此外，所有模型都是简化的。如果技术发生变化，例如互联网减少了对实体城市集中化的需求，那么恒定承载容量K的假设可能不再成立。
9. Case Study: Modelling China's Urbanisation Since 1978 | 案例分析：1978年以来中国的城市化建模
China's urbanisation share jumped from 17.9% in 1978 to over 64% by 2020. Fitting a logistic curve to this data gives an estimated K of approximately 75–80%, with an inflection point around 2008. The growth rate r turned out to be about 0.045 per year. The model suggests that China entered the deceleration phase after 2020, aligning with current trends.
中国的城市化比例从1978年的17.9%跃升至2020年的64%以上。将逻辑斯蒂曲线拟合到这些数据上，得到估计的K约为75–80%，拐点大约在2008年。增长率r约为每年0.045。该模型表明中国在2020年后已进入减速阶段，这与当前趋势相符。
Using the logistic solution U(t) = K / [1 + ((K−U₀)/U₀) e⁻ʳᵗ], we can predict the year when China will reach 70% . Solving the equation yields around 2030, though this depends on the accuracy of K and r.
使用逻辑斯蒂解 U(t) = K / [1 + ((K−U₀)/U₀) e⁻ʳᵗ]，我们可以预测中国达到70%城市化比例的年份。解方程得出大约在2030年，不过这取决于K和r的准确性。
10. Comparative Regression Across Regions | 跨区域回归比较
By applying the same logistic regression to different continents, we can compare the intrinsic growth rates r and carrying capacities K. The table below summarises estimates based on UN data from 1950 to 2020:
| Region | r (per year) | K (urban share) | R² |
|---|---|---|---|
| Sub-Saharan Africa | 0.032 | 0.62 | 0.987 |
| Latin America | 0.041 | 0.84 | 0.992 |
| Europe | 0.029 | 0.77 | 0.979 |
Africa's low K suggests a longer road to urbanisation, while Latin America's higher K reflects an already heavily urbanised region. These parameters can be discussed in the context of historical migration and economic development.
非洲较低的K值暗示其城市化道路更长，而拉丁美洲较高的K值则反映该地区已经高度城市化。这些参数可以在历史迁移和经济发展的背景下进行讨论。
11. Future Projections and Urban Sustainability | 未来预测与城市可持续性
Extrapolating mathematical models gives planners forecasts to prepare for housing, transport, and energy demands. However, the logistic model assumes a symmetric S-curve, which may not hold if urbanisation slows more quickly at high levels due to counter-urbanisation. Modified models, such as the Richards' growth curve, introduce a shape parameter to allow asymmetric transitions.
通过对数学模型进行外推，规划者可以做出预测，为住房、交通和能源需求做准备。然而，逻辑斯蒂模型假设对称的S曲线，但如果由于反城市化导致高水平城市化减速更快，该假设可能不成立。修改后的模型，如Richards增长曲线，引入了一个形状参数以允许非对称过渡。
may also model urban density gradients using negative exponential functions: D(x) = D₀ e⁻ᵇˣ e city centre. This connects to integral calculus when determining total urban population.
在数学中，我们还可以用负指数函数模拟城市密度梯度：D(x) = D₀ e⁻ᵇˣ，其中D是距离市中心x处的人口密度。在确定城市总人口时，这会与积分相关联。
These advanced topics bridge A-Level mathematics with university-level urban geography, demonstrating the interdisciplinary value of mathematical modelling.
这些高级主题将A-Level数学与大学水平的城市地理学联系起来，展示了数学建模的跨学科价值。
12. Conclusion: The Power of Mathematical Analysis | 结论：数学分析的力量
The patterns of urbanisation since 1945 are not merely historical narratives but rich datasets that invite mathematical exploration. From exponential and logistic growth to differential equations and regression analysis, the techniques learned in Edexcel A-Level Mathematics enable students to decode complex real-world phenomena. By modelling urbanisation, we appreciate how rates of change, equilibrium, and data transformation form the very language of demographic change. The next time you see a skyline, remember the equations that predict its expansion.
1945年以来的城市化模式不仅仅是历史叙事，更是丰富的数据集，等待着数学探索。从指数增长和逻辑斯蒂增长到微分方程和回归分析，Edexcel A-Level数学中学到的技术使学生能够解码复杂的现实世界现象。通过对城市化进行建模，我们体会到变化率、平衡态和数据转换如何构成了人口变化的语言。下次你看到天际线时，请记住那些预测其扩张的方程。
Published by TutorHao | Mathematics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图