Business Intelligence (BI) is a set of methodologies and resources (theoretical concepts, algorithms, software, tools, technologies) used in business environments and whose fundamental objective is to transform information into knowledge. The purpose of Business Intelligence is to support better business decision making to improve the productivity and performance of any business organization.
商业智能(BI)是在商业环境中使用的一组方法论和资源(理论概念，算法，软件，工具，技术)，其基本目标是将信息转化为知识。 商业智能的目的是支持更好的商业决策，以提高任何商业组织的生产力和绩效。
BI relies heavily on a core set of analytics tools: Balanced Scorecards, Dashboards, and Key Performance Indicators.
BI在很大程度上依赖于一组核心分析工具：平衡计分卡，仪表板和关键绩效指标。
The Balanced Scorecard (BS) is a management tool that allows the integration of short-term operational control with the long-term vision and strategy of the company under study. To this purpose, the BS formulates, implements, controls, and manages the organization’s strategy, efficiently linking strategic planning, operational management, and evaluation of group and individual performances.
平衡计分卡(BS)是一种管理工具，可将短期运营控制与所研究公司的长期愿景和战略相集成。 为此，BS制定，实施，控制和管理组织的战略，有效地将战略计划，运营管理以及对团队和个人绩效的评估联系在一起。
Key Performance Indicators (KPIs) are the most important metrics that any organization needs to measure, monitor, and perform against in order to reach its strategic plan.
关键绩效指标(KPI)是任何组织为了达成其战略计划而需要衡量，监控和执行的最重要指标。
A Business Intelligence Dashboard is a tech-driven approach that uses data visualization tools to display KPIs in order to track the main factors that contribute to the operation and success of the business. It is a graphic representation of the main metrics involved in the pursuit of the objectives set by the company.
商业智能仪表板是一种技术驱动的方法，使用数据可视化工具显示KPI，以便跟踪有助于业务运营和成功的主要因素。 它是实现公司设定目标所涉及的主要指标的图形表示。
Business owners and corporate executives always want simple solutions. In that sense, data analysts must make dashboards easy to look at, navigate and understand. Two traditional graphics used to display KPIs on dashboards are Gauge Charts and Bullet Charts.
企业所有者和企业高管始终希望获得简单的解决方案。 从这个意义上讲，数据分析师必须使仪表板易于查看，浏览和理解。 用于在仪表板上显示KPI的两个传统图形是仪表图和项目符号图。
Let’s analyze them separately.
让我们分别分析它们。
1.-量规表 (1.- Gauge Charts)
AKA: Speedometer charts, Dial charts, Angular Gauge charts
AKA ：里程表，转盘图，角规图
WHY: widely used in Business Intelligence (BI) visualizations, particularly in dashboards, to indicate whether a data or measurement is within, below, or above a certain range of values. They are very simple graphics that manage to capture the attention of the audience very quickly, even the non-specialized one. They provide information about a single quantitative measurement, comparing its present value to a goal and to a series of ranges indicated by bands of different colors. The following figure indicates that the current sales value (420) is within a satisfactory range but 30 units below the predetermined goal.
为什么：广泛用于商业智能(BI)可视化中，尤其是在仪表板中，以指示数据或度量值是在某个值范围之内，之下还是之上。 它们是非常简单的图形，可以非常Swift地吸引观众的注意力，即使是非专业的图形也是如此。 他们提供有关单个定量测量的信息，将其当前值与目标以及由不同颜色的条带指示的一系列范围进行比较。 下图显示当前销售价值(420)在令人满意的范围内，但比预定目标低30个单位。
HOW: Although there are different formats, the most common corresponds to the speedometer of a vehicle: a radial numerical scale separated into sectors, each identified by a particular color; a pointer or a needle moving over the radial scale; a minimum or lower limit value, a maximum or upper limit value and a target value. The number of ranges or sectors is variable, but usually there are three sectors, identified by a red color the sector that is below the lower threshold, a yellow color between thresholds, and a green one the sector that is above the upper threshold. It is not recommended to display more than five ranges or sectors.
HOW ：尽管有不同的格式，但最常见的格式对应于车辆的速度表：径向数字刻度分为多个扇区，每个扇区由一种特定的颜色标识； 在径向刻度上移动的指针或指针； 最小或下限值，最大或上限值以及目标值。 范围或扇区的数量是可变的，但通常有三个扇区，用红色标识低于下限阈值的扇区，用黄色标识阈值之间的颜色，用绿色标识高于上限阈值的扇区。 建议不要显示五个以上的范围或扇区。
The following are some of the possible qualitative values associated with the different colors: poor, satisfactory, and good; green for satisfactory, yellow for caution, and red for alarm; poor, average, and excellent; red for low performance, yellow for normal performance, and green for high performance. These ranges of qualitative values are widely used in a BI presentation or on a dashboard to show the business performance of a given company. The target value can be an objective to be achieved, a benchmark or a previous value to be exceeded.
以下是与不同颜色相关的一些可能的定性值：差，令人满意和良好； 绿色表示满意，黄色表示谨慎，红色表示报警； 差，中等和优秀； 红色表示低性能，黄色表示正常性能，绿色表示高性能。 这些定性值范围广泛用于BI演示或仪表板中，以显示给定公司的业务绩效。 目标值可以是要达到的目标，基准或要超过的先前值。
Do not confuse gauges charts with donut graphs (Doughnut charts). While the former only reports a quantitative measurement, donut charts are related to a whole that can be divided into individual parts and how each part is related -in a relative or absolute way- with the total amount (Part of a whole analysis).
不要将仪表盘图表与甜甜圈图(甜甜圈图)混淆。 尽管前者仅报告定量测量，但甜甜圈图与整体相关，可以将其划分为各个部分，以及每个部分如何相对(以相对或绝对方式)与总量(整个分析的一部分)相关。
Gauge charts have several virtues when telling a story with data:
在用数据讲故事时，量规图具有多种优点：
· They are very simple, easy to understand, and quickly capture the attention of the audience;
·它们非常简单，易于理解，并Swift吸引了观众的注意力；
· Audiences are familiar with them: car speedometers, indicators in computer games, temperature indication in ovens or stoves, etc. Unlike more complex graphics, there is no need to “waste time” in explaining them;
·观众熟悉它们：汽车速度表，计算机游戏中的指示器，烤箱或火炉中的温度指示等。与更复杂的图形不同，不需要在解释它们时“浪费时间”；
· The sequence of colors, from red to green, is intuitive and immediate to understand;
·从红色到绿色的颜色顺序直观且易于理解；
· Most business and corporate executives prefer them because they transmit key data simply, clearly, and immediately;
·大多数企业和公司高管更喜欢它们，因为它们可以简单，清晰，立即地传输关键数据；
But they are also riddled with disadvantages, which must be taken into account at the moment of the storytelling:
但是它们也充满弊端，在讲故事时必须考虑到这些弊端：
· Their simplicity does not allow them to portray the context;
·它们的简单性不允许他们描述上下文；
· They often mislead the audience by omitting key information, resulting in poor storytelling.
·他们常常通过省略关键信息来误导观众，导致故事叙述不佳。
· Humans have difficulty comparing numerical values by means of angles; we always prefer to make comparisons by means of lengths starting from a common baseline (bullet charts);
·人们很难通过角度比较数值； 我们始终喜欢通过从一个共同的基线(项目符号图表)开始的长度进行比较；
· They take up a lot of physical space and their format can be distracting to the audience. Dashboards and Scorecards are often cluttered with indicators, which makes the storytelling process much more complicated;
·它们占用了大量的物理空间，其格式可能会分散观众的注意力。 仪表板和记分卡上通常都杂乱着指示器，这使讲故事的过程变得更加复杂。
· Remember that color blindness (color vision deficiency) affects approximately one in twelve men and one in two hundred women. The most common and problematic form of color deficiency is called “red-green color blindness,” precisely two of the colors commonly used in gauges. For these reasons, the actual trend is to encode the sectors as a single hue with distinct intensities ranging from dark to light. Darker colors intensities are used for low or poor values, while lighter colors intensities indicate satisfactory or excellent values (Fig. 2);
·请记住，色盲(色盲)会影响大约十二分之一的男性和一百分之二的女性。 色差最常见且最成问题的形式称为“红绿色色盲”，恰好是量规中常用的两种颜色。 由于这些原因，实际趋势是将扇区编码为具有从暗到亮的不同强度的单个色调。 较深的颜色强度用于较低或较差的值，而较浅的颜色强度表示令人满意或极好的值(图2)。
The last warning comes from Robert Kosara (#1): “In data visualization, gauges simply solve the wrong problem. Business decisions cannot be made only on the current value of some measure, but need history and context. Values also typically do not change as fast, nor are as directly coupled, as the amount of pressure on the accelerator and the car’s speed.”
最后的警告来自Robert Kosara(＃1)：“在数据可视化中，仪表可以解决错误的问题。 业务决策不能仅根据某种衡量标准的当前价值来做出，而需要历史和上下文。 值通常也不会像加速器上的压力大小和汽车的速度变化一样快，也没有直接耦合。”
The Matplotlib library does not have a function that allows us to draw a gauge chart directly. A very ingenious solution to the problem can be found on the following website: https://waterprogramming.wordpress.com/2018/06/04/creating-shaded-dial-plots-in-python/. A simpler solution is based on the Plotly graphic library which has an Indicator function: https://plotly.com/python/gauge-charts/.
Matplotlib库没有允许我们直接绘制仪表图的功能。 可以在以下网站上找到非常巧妙的解决方案： https : //waterprogramming.wordpress.com/2018/06/04/creating-shaded-dial-plots-in-python/ 一个更简单的解决方案基于具有指示器功能的Plotly图形库： https : //plotly.com/python/gauge-charts/ 。
2.-项目符号图 (2.- Bullet Charts)
AKA: Bullet Graphs
AKA ：项目符号图
WHY: Renowned data visualization expert Stephen Few developed an alternative to gauges that he named Bullet Chart (#3). The key idea was to enrich the information by adding complementary measures to portray the context, using less physical space, and without decorations that distract the attention of the audience. Bullet charts provide information about a single quantitative measurement, comparing its present value against a target and to a series of ranges indicated by bands of different colors.
为何：著名的数据可视化专家Stephen Few开发了一种替代量规的方法，他将其命名为Bullet Chart(＃3)。 关键思想是通过添加补充性措施来描绘背景来丰富信息，使用较少的物理空间并且不进行会分散观众注意力的装饰。 项目符号图提供有关单个定量测量的信息，将其当前值与目标值以及由不同颜色的条带指示的一系列范围进行比较。
HOW: they are similar to standard bar charts (they code information by length or height) but differ from them in that they include a central narrow bar that indicates the present (current) value of the numerical variable being reported. They also have a perpendicular line that shows the target or objective, and different ranges or sectors (usually three) that are indicated with bands of different colors or with varying intensities of a single hue. Unlike gauge charts, different categories can be compared in a single diagram.
HOW ：它们与标准条形图相似(它们的长度或高度代表代码信息)，但与它们不同之处在于它们包括一个中央窄条，用于指示要报告的数字变量的当前(当前)值。 它们还具有一条垂直线，显示目标或物体，以及不同的范围或扇区(通常为三个)，这些范围或扇区用不同颜色的条带或单个色调的不同强度指示。 与仪表图不同，可以在一个图中比较不同类别。
The standard bullet chart has five components (#3):
标准项目符号图表包含五个组件(＃3)：
· a text label indicating the measured variable and the corresponding units of measurement;
·文本标签，指示测量变量和相应的测量单位；
· a quantitative scale along a linear axis with its ticks and numbers identifying equi-spaced measurement intervals;
·沿线性轴的定量刻度，其刻度和数字标识等距的测量间隔；
· a primary or featured measure encoded with a clearly visible bar and which may have a baseline other than zero;
·用清晰可见的条编码的主要或特色量度，其基线可能不为零；
· comparison measures in the form of thin lines drawn perpendicular to the chart orientation (vertical lines if the bar representing the primary measure is horizontal, horizontal lines if the bar representing the primary measure is vertical);
·比较措施，采取与图表方向垂直的细线形式(如果代表主要度量的条形为水平，则为垂直线；代表主要度量的条形为垂直，则为水平线)；
· two to five qualitative ranges of performance that indicate the qualitative status of the primary measure. The ranges are differentiated by distinct colors or varying intensities of a single hue.
·2到5个定性绩效范围，表明主要措施的定性状态。 范围通过不同的颜色或单个色调的变化强度来区分。
Bullet charts are presented in an either vertical or horizontal orientation. A horizontal bullet chart allows different categories to be displayed on a vertical axis while the numerical values are indicated by a quantitative scale on the horizontal axis. Reciprocally, a vertical bullet chart allows different categories to be displayed on a horizontal axis while the numerical values are shown by a quantitative scale on the vertical axis. The quantitative scale allows you to display the value of the variable measured on a linear axis, easier to read than the angular axis of the gauges.
项目符号图以垂直或水平方向显示。 水平项目符号图允许在垂直轴上显示不同的类别，而数值则通过水平轴上的定量标度表示。 相反，垂直项目符号图允许在水平轴上显示不同的类别，而数值在垂直轴上以定量刻度显示。 定量刻度允许您显示在线性轴上测得的变量的值，比量规的角轴更易于阅读。
Similar to gauge charts, the most usual qualitative values of the ranges or sectors are: poor, satisfactory, and good; green for satisfactory, yellow for caution, and red for alarm (Fig. 4); poor, average, and excellent; below performance, average performance, and over performance.
与轨距图相似，范围或扇区的最常见定性值为：差，令人满意和良好； 绿色表示满意，黄色表示谨慎，红色表示报警(图4)； 差，中等和优秀； 低于性能，平均性能和高于性能。
The numerical variables usually indicated by bullet charts on dashboards and other BI visualizations are Sales, Income, Profit, Expenses, Number of Customers, and related.
仪表板和其他BI可视化文件上的项目符号图通常指示的数字变量是“销售”，“收入”，“利润”，“费用”，“客户数量”及相关。
Do not confuse bullet charts with progress bar charts. The latter is a graphical control element used to visualize the progress of a certain operation such as downloading a file, transferring files, or installing software. Some visualization tools suggest the use of progress bars as business indicators as shown in the figure below:
不要混淆进度条形图子弹图。 后者是图形控制元素，用于可视化某些操作的进度，例如下载文件，传输文件或安装软件。 一些可视化工具建议使用进度条作为业务指标，如下图所示：
As with the gauge chart, Matplotlib does not have a function that allows you to draw a bullet chart immediately. Similarly, I found a very ingenious solution based on Matplotlib and Seaborn: https://pbpython.com/bullet-graph.html. I also found a better solution based on Plotly: https://plotly.com/python/bullet-charts/
与仪表图一样，Matplotlib也没有允许您立即绘制项目符号图的功能。 同样，我发现了一个基于Matplotlib和Seaborn的非常巧妙的解决方案： https ://pbpython.com/bullet-graph.html。 我还发现了基于Plotly的更好的解决方案： https ://plotly.com/python/bullet-charts/
A scientific study conducted at a Chinese university compared the emotional differences of 40 college students when confronted with horizontal bullet charts, vertical bullet charts, and angular gauge charts (#4). The cognitive load demanded by each one of the graphic alternatives, the level of satisfaction perceived by the students, and the aesthetic effect of each one of the visualization alternatives were used as subjective metrics. The study concluded that vertically oriented bullet charts generated the highest satisfaction and lowest cognitive load. In contrast, angular gauge charts resulted in the lowest satisfaction, highest cognitive load, although they were the most aesthetically pleasing.
在中国大学进行的一项科学研究比较了40名大学生在面对水平子弹图，垂直子弹图和角规图时的情感差异(＃4)。 每个图形替代方案所要求的认知负荷，学生所感知的满意度水平以及每个可视化替代方案的美学效果均用作主观指标。 研究得出结论，垂直定向的项目符号图产生了最高的满意度和最低的认知负荷。 相比之下，尽管角度规图表在美学上是最令人愉悦的，但却导致满意度最低，认知负荷最高。
I recommend you to use these results for your next business presentation.
我建议您将这些结果用于下一个业务演示。
In the context of the digital transformation of many companies, Business Intelligence and Data Visualization techniques are becoming increasingly relevant. The spectrum of visualization and chart customization possibilities is overwhelming: there are dozens of different graphs that provide great flexibility when visualizing data, but sometimes only simple, easy-to-understand diagrams that quickly capture the attention of the audience are required. Gauges and bullet charts provide that simplicity and familiarity, which is much appreciated by audiences.
在许多公司进行数字化转型的背景下，商业智能和数据可视化技术变得越来越重要。 可视化和图表自定义的可能性种类繁多：数十种不同的图形在可视化数据时提供了极大的灵活性，但有时只需要简单，易于理解的图表即可Swift吸引观众的注意力。 量表和项目符号图提供了这种简单性和熟悉性，深受观众赞赏。
Particularly, we suggest the use of bullet charts for their ability to encode more information in smaller spaces.
特别是，我们建议使用项目符号图表来在较小的空间中编码更多信息。
If you find this article of interest, please read my previous:
如果您发现这篇文章感兴趣，请阅读我以前的文章：
“Histograms, Why & How, Storytelling, Tips & Extensions”
“直方图，为什么和如何，讲故事，技巧和扩展”
“Parallel Coordinates Plots, Why & How, Storytelling with Parallels”
“平行坐标图，为什么和如何，讲故事与平行”
#1: https://eagereyes.org/criticism/data-display-vs-data-visualization
＃1： https ： //eagereyes.org/criticism/data-display-vs-data-visualization
＃2： https ： //vizzlo.com/
#3: https://www.perceptualedge.com/articles/misc/Bullet_Graph_Design_Spec.pdf
＃3： https ： //www.perceptualedge.com/articles/misc/Bullet_Graph_Design_Spec.pdf
#4: Lei Wu, Lingli Guo, Hao Fang, and Lijun Mou, “Bullet Graph Versus Gauges Graph: Evaluation Human Information Processing of Industrial Visualization Based on Eye-Tracking Methods”, T. Z. Ahram and C. Falcão (Eds.): AHFE 2018, AISC 794, pp. 752–762, 2019. https://doi.org/10.1007/978-3-319-94947-5_74
＃4：吴磊，郭凌丽，方浩和牟立军，“子弹图与量表图：基于眼动跟踪方法评估工业可视化的人类信息处理”，TZ Ahram和C.Falcão(编辑)：AHFE 2018年，AISC 794，第752–762页，2019年.https： //doi.org/10.1007/978-3-319-94947-5_74
翻译自: https://towardsdatascience.com/gauge-bullet-charts-cfe171ca3094
6187
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
