工作日各时间段的叫车量、等待时间、调度时间

工作日各时间段的叫车量、等待时间、调度时间

问题：统计周一到周五各时段的叫车量、平均等待接单时间和平均调度时间。
全部以event_time-开始打车时间为时段划分依据，平均等待接单时间和平均调度时间均保留1位小数，平均调度时间仅计算完成了的订单，结果按叫车量升序排序。

注：
不同时段定义：早高峰 [07:00:00 , 09:00:00)、工作时间 [09:00:00 , 17:00:00）、晚高峰 [17:00:00 , 20:00:00）、休息时间 [20:00:00 , 07:00:00）
时间区间左闭右开（即7:00:00算作早高峰，而9:00:00不算做早高峰）
从开始打车到司机接单为等待接单时间，从司机接单到上车为调度时间。

1- 数据准备

表1: 用户打车信息表:t_taxi_info
id(主键id), uuid(用户id), order_id(订单id), city_name(城市), event_time(开始打车时间), end_time(结束打车时间:接单/订单取消)

表2: 订单信息表:t_order_info
id(主键id), uuid(用户id), order_id(订单id), driver_id(司机id), order_time(接单时间), start_time(订单开始时间), finish_time(订单完成时间), fare(费用), grade(评分)

注: 两张表的order_id 都是唯一值

-- 数据准备
WITH t_taxi_info AS (
SELECT * FROM (
  VALUES      (1, 1001, 90004, '深圳市', '2023-12-04 11:00:12', '2023-12-04 11:00:42')
            , (2, 1002, 90011, '深圳市', '2023-12-05 21:00:22', '2023-12-05 21:01:06')
            , (3, 1002, 90008, '深圳市', '2023-12-05 19:01:02', '2023-12-05 19:01:52')
            , (4, 1003, 90006, '深圳市', '2023-12-06 09:01:06', '2023-12-06 09:01:36')
            , (5, 1004, 90012, '深圳市', '2023-12-06 18:01:08', '2023-12-06 18:01:48')
            , (6, 1004, 90018, '深圳市', '2023-12-07 08:01:16', '2023-12-07 08:01:43')
            , (7, 1006, 90023, '深圳市', '2023-12-07 08:01:23', '2023-12-07 08:02:01')
            , (8, 1007, 90002, '深圳市', '2023-12-08 20:01:12', '2023-12-08 20:01:39')
            , (9, 1006, 90017, '深圳市', '2023-12-08 08:30:18', '2023-12-08 08:30:36')
) AS table_name(id, uuid, order_id, city_name, event_time, end_time)
)

, t_order_info AS (
SELECT * FROM (
  VALUES      (1, 1001, 90004, 2002, '2023-12-04 11:00:42', '2023-12-04 11:05:22', '2023-12-04 11:40:41', 38, 4.8)
            , (2, 1002, 90011, 2004, '2023-12-05 21:01:06', '2023-12-05 21:04:02', '2023-12-05 21:52:52', 46, 4.6)
            , (3, 1002, 90008, 2005, '2023-12-05 19:01:52', '2023-12-05 19:08:52', '2023-12-05 19:33:20', 29, 4.3)
            , (4, 1003, 90006, 2004, '2023-12-06 09:01:36', '2023-12-06 09:03:36', '2023-12-06 09:36:30', 30, 4.5)
            , (5, 1004, 90012, 2008, '2023-12-06 18:01:48', '2023-12-06 18:02:48', '2023-12-06 18:42:20', 46, 4.7)
            , (6, 1004, 90018, 2004, '2023-12-07 08:01:43', '2023-12-07 08:04:46', '2023-12-07 08:53:41', 52, 4.8)
            , (7, 1007, 90002, 2021, '2023-12-08 20:01:39', '2023-12-08 20:06:33', '2023-12-08 20:29:51', 42, 4.9)
            , (8, 1006, 90017, 2016, '2023-12-08 08:30:36', '2023-12-08 08:34:32', '2023-12-08 09:12:21', 48, 4.1)
) AS table_name(id, uuid, order_id, driver_id, order_time, start_time, finish_time, fare, grade)
)

2- 代码实现

-- 1. 数据处理: 获取等待时间,调度时间, 时间类型
SELECT 
      a.order_id
    , a.event_time
    , a.end_time
    , b.order_time
    , b.start_time
    , (UNIX_TIMESTAMP(a.end_time) - UNIX_TIMESTAMP(a.event_time) ) AS wait_times -- 等待时间
    , (UNIX_TIMESTAMP(b.start_time) - UNIX_TIMESTAMP(b.order_time)) AS schedulers_times  -- 调度时间
    , (CASE
         WHEN HOUR(a.event_time) >= '07' AND HOUR(a.event_time) < '09' THEN "早高峰"
         WHEN HOUR(a.event_time) >= '09' AND HOUR(a.event_time) < '17' THEN "工作时间"
         WHEN HOUR(a.event_time) >= '17' AND HOUR(a.event_time) < '20' THEN "晚高峰"
         ELSE "休息时间" END ) as time_type
FROM t_taxi_info a 
JOIN t_order_info b 
ON a.order_id = b.order_id
;

order_id	event_time	end_time	order_time	start_time	wait_times	schedulers_times	time_type
90004	2023-12-04 11:00:12	2023-12-04 11:00:42	2023-12-04 11:00:42	2023-12-04 11:05:22	30	280	工作时间
90011	2023-12-05 21:00:22	2023-12-05 21:01:06	2023-12-05 21:01:06	2023-12-05 21:04:02	44	176	休息时间
90008	2023-12-05 19:01:02	2023-12-05 19:01:52	2023-12-05 19:01:52	2023-12-05 19:08:52	50	420	晚高峰
90006	2023-12-06 09:01:06	2023-12-06 09:01:36	2023-12-06 09:01:36	2023-12-06 09:03:36	30	120	工作时间
90012	2023-12-06 18:01:08	2023-12-06 18:01:48	2023-12-06 18:01:48	2023-12-06 18:02:48	40	60	晚高峰
90018	2023-12-07 08:01:16	2023-12-07 08:01:43	2023-12-07 08:01:43	2023-12-07 08:04:46	27	183	早高峰
90002	2023-12-08 20:01:12	2023-12-08 20:01:39	2023-12-08 20:01:39	2023-12-08 20:06:33	27	294	休息时间
90017	2023-12-08 08:30:18	2023-12-08 08:30:36	2023-12-08 08:30:36	2023-12-08 08:34:32	18	236	早高峰

-- 2. 计算各时间段的叫车量,等待时间平均值,调度时间平均值
SELECT 
      time_type
    , COUNT(order_id) AS get_taxi_num
    , SUM(wait_times)/COUNT(order_id) AS wait_times_avg
    , SUM(schedulers_times)/COUNT(order_id) AS schedulers_times_avg
FROM (
    SELECT 
          a.order_id
        , a.event_time
        , a.end_time
        , b.order_time
        , b.start_time
        , (UNIX_TIMESTAMP(a.end_time) - UNIX_TIMESTAMP(a.event_time) ) AS wait_times -- 等待时间
        , (UNIX_TIMESTAMP(b.start_time) - UNIX_TIMESTAMP(b.order_time)) AS schedulers_times  -- 调度时间
        , (CASE
             WHEN HOUR(a.event_time) >= '07' AND HOUR(a.event_time) < '09' THEN "早高峰"
             WHEN HOUR(a.event_time) >= '09' AND HOUR(a.event_time) < '17' THEN "工作时间"
             WHEN HOUR(a.event_time) >= '17' AND HOUR(a.event_time) < '20' THEN "晚高峰"
             ELSE "休息时间" END ) as time_type
    FROM t_taxi_info a 
    JOIN t_order_info b 
    ON a.order_id = b.order_id
) t1
GROUP BY time_type
;

time_type	get_taxi_num	wait_times_avg	schedulers_times_avg
休息时间	2	35.5	235.0
工作时间	2	30.0	200.0
早高峰	2	22.5	209.5
晚高峰	2	45.0	240.0

end