Papers Category:Science Papers
- Statistics Papers
Post Time:2011-9-7 10:13:00
Abstract: In recent years, dynamics of human behavior on a large number of studies have shown that human activities close to the statistical distribution of interval power-law distribution through the sales order processing behavior and distribution of time intervals, indicating that the same power-law function with a similar distribution But different types of sales are not the same pattern of behavior is only through continuous statistical study can be found a variety of sales practices laws of this article from the human point of view to analyze the dynamics of group activities and annual sales order processing, an attempt to find the company's sales practices in the RA kinetics. The study found that sales of each project tracking time interval distribution and delivery of each order interval distributions are subject to power-law distribution.
Keywords: Human Kinetics sale order processing power-law distribution
The complexity of human behavior research, has from the earliest of human Huodong summarized as steady, or completely random, Poisson process described can be used, transferred to the Ren Lei Dong Lixue model-based studies. Through a lot of evidence that human activity is quite uneven in these acts, generally found to have similar characteristics deviate from the Poisson process of these phenomena demonstrate that in addition strongly influenced by the menstrual cycles act the part, the time interval of non-Poisson statistics shows characteristics of is heavy-tailed distribution, may be prevalent in human behavior, while this distribution function has a similar power-law curve.
With in-depth study of complex systems that deal with human behavior in the e-mail and written communication activities are consistent with the typical power-law distribution P (t) = ta, where a 1.0 and 1.5 respectively, thus the human dynamics is divided into two universality class, their characteristics scaling exponent is 1.0 and 1.5, respectively. Benwen RA Gongsi statistics of the branch 13 of a project to track sales activity behavior and sales order processing situation, trying to existing models of human dynamics and the actual of control. Study the daily and annual sales practices laws of order processing, but also for future research in this direction a little example.
2 Data source description This data comes from the company's sales order data and daily activities program.
RA task of the company's sales average of about one million U.S. dollars in the 1.4-1.5, regular clients and track the number of items per order amount of sales in the online records are required each months should be reviewed for its accuracy compared, while actual sales have all entered into the database, you can directly query the database based on the number of orders, amount and date, an objective response to the annual processing of orders.
According to the daily activities of sales can be seen, the type of analysis of customer sales activities mainly in the type of customer and project-based OEM customer oriented and each time the customer to track project information from creation to win (or lose, the delay between the time time interval of a project because the data is updated once a month for the general requirements, so the statistics, the time interval unit for 30 days is reasonable, according to the general law of sales, OEM customers have the customer order is fixed, the time interval is generally a - 3 months, more laws, greater grasp of the project-based customers to track time more than half the time usually only order generation, winning percentage is generally between 40% -60%. The data used for the 07 to 10 completed the sale of a total of 470 times. Sales order processing in general accordance with the provisions of the order cycle, the company 4-8 weeks to process.
Of course there are some quite sure of the order, request a tight schedule, you can not order in advance to deal with the logistics department stocking done in can pick up immediately after signing the contract, so in order processing statistics last week (7 days for order processing unit to the time interval statistics. The data used for the 08 years to complete the order processing statistics, a total of 393 times.
3 Statistical analysis
3.1 customer sales activity patterns of statistical
For each sales engineers, project progress and what time the customer orders by the sales engineer can determine the subjective desire, even fixed OEM customers, there are uncertainties in the company's products have been identified using the RA, it includes many uncontrollable specific questions: What time can be such as project approval, project finance operations, changes in market demand is, therefore, found that the project engineer for each sales opportunity to receive the order this time interval could not help entirely personal decision, but all sales engineers through statistical time interval for each client activity approach to the analysis, we find the general rule.
Harvest time t with the order that found the project to receive orders from the time interval, an order received on day t can be approximated by the probability of a power function P (t) = ta, in which a = 1.34. Of which 62.55% of orders are received within 12 months Over 24 months of orders less than 8%, which is also in line with company sales tasks. Links to free download http://www.hi138. com
3.2 Sales Order Processing laws of statistics
For each order received, starting from the signing of the order, all delivery to the customer, this process depends entirely on the client's financial situation and the actual progress of the project requirements, the seller can not pick up about the situation. Likewise, the statistical methods to analyze all the entire year Orders can also be found in the general law.
To t that the contract to the time interval between delivery, an order received on day t can be approximated by the probability of a power function P (t) = ta, in which a = 1.62. 80% of orders for delivery in 8 weeks , is also consistent with the requirements of the company operating funds of which 24 weeks after the delivery of orders less than 4%, which also shows the authenticity of the order and sales personnel to grasp the extent of the customer.
These two simple models to meet this statistical distribution of interval sales activities in customer tracking and order delivery time interval distribution of two events, can be found from a statistical point of sale and order processing dynamics of human behavior both power-law model is similar to the time distribution of this study's findings is consistent with the dynamics of human activity model sales order processing activities and behavior of the distribution of two indices were 1.34 and 1.62 of this data with the distribution of index close to 1.5. Through this research, hope for the future quantitative analysis of the statistical characteristics of human behavior to provide some actual cases, but also for the scientific management of sales activities and order processing to bring new management perspective. ◆
 Wang Binghong, human behavior, dynamics and statistical mechanics study ,2009-06
 Li Nannan, Dengzhu Jun, Zhang Ning, Li Jimin, Qin Yan, Zhou Tao from Qian mode of communication to see the scaling law of human dynamics, 2008  Joao Gama Oliveira, Albert-Laszlo Barabasi, Darwin and Einstein mode of communication, 2005 Links to free download http://eng.