CV
Biografy
Abstract

Theme: Automated subsystem of planning and coordinating the work of commercial agents


Aims and objectives

The company sells its production to retail outlets (RO). For this purpose, with certain intervals of time commercial agent (CA) visits RO, that he/she is responsible for. Each CA is responsible for 60-100 RO. Each CA is subordinate to his / her supervisor - sales manager (SM), whose task is to define a set of RO, which CA has to visit during the week. About 8-15 CA are under control of one SM. Lets denote a set of RO as ij and a set of CA as Ci. The general scheme of work on present day looks like that:

  1. .SM makes a daily schedule of visiting retail outlets for CA, depending on several factors: the frequency of visiting RO, the importance of visiting on a certain day of the week, the importance of attendance at a particular time of day, etc. In a single day a limited number of RO can be visited.
  2. After making the schedule for the CA, an existing subsystem determines the optimal route of visiting RO for every day.

RO is characterized by a set of parameters: Tij{P,Q,t}.

Р – frequency of visiting RO.

Q – the importance of visiting the RO at a certain time interval.

t – time of a day for visiting RO.

Subsystem of scheduling the visits to RO can be represented by the following scheme.


The general scheme of subsystem’s work

Illustration 1. The general scheme of subsystem’s work


Topicality of the work

The solution of the problem of scheduling depends on the problem of determining the list of RO by the following parameters: the time of a visit RO - tij , frequency of visiting RO - Pij. Depending on these parameters the number of visits per day can vary because of the changes in the length of the done path. Therefore the topicality of the work lies in identifying the optimal sets of RO for visiting each day of the week. Thus SM has to make sets L{D,T,C}
D – day of the week.
Т – set of RO for visiting.
С – CA.
This will allow to implement the feedback in the existing system and improve the efficiency of the making schedule’s subsystem. After upgrading the subsystem can be represented in a following way:


The general scheme of subsystem’s work after upgrading

Illustration 2. The general scheme of subsystem’s work after upgrading


Thus at the stage of forming a set of RO for visiting L{D,T,C,K} where K – the route’s efficiency factor, SM will be able to take into account the schedule of visiting RO.

In this case an already existing subsystem of determining the optimal route by selected points can be the fitness function.

Formula

W – number of working days in a week for the CA.

S – the function of determining the length of optimal route by selected RO.


Accounting of possible routes and their effectiveness on the stage of forming a set of RO will let increase the efficiency factor of CA, and take into account the requirements for visiting RO, which ultimately will enhance the company's earnings.

The main remarkable aspect is the connection of making the optimal schedule’s system with the subsystem of finding the optimal route.
Currently, there are many systems which solve similar tasks. They usually focused on the logistics of transport. However, the majority draw up a route only on certain period of time, such as working hours. Also these systems do not take into account the individual specifics of enterprise activity, such as the importance of visiting RO at a specific time of a day, etc. Thus, we see a strong need of further development and research in the field of scientific activity.


Conclusion

At the moment the integration of the subsystem with the corporate database is the most likely direction of development of such systems. This will increase the number of factors taken into account and improve the efficiency of scheduling of visiting retail outlets by commercial agents.


REFERENCES

  1. Скобцов Ю.А. «Основы эволюционных вычислений».— Учебное пособие.— Донецк: ДонНТУ, 2008. — С. 326.
  2. Седжвик Р. Фундаментальные алгоритмы [Текст]. – Киев: Азимут-Украина, 2003. – С. 154-190.
  3. Майника Э. — Алгоритмы оптимизации на сетях и графах [Текст]. — Санкт-Питербург: Unipress, 2009. — С. 91-115.
  4. Beck, Ken — Extreme Programming Explained: Embrace Change [Text] / [Текст]. Atlanta, GA: Cokesbury Book, 2009. — C. 50-55.
  5. Мудров В.И. Задача о коммивояжере. — М.: «Знание», 1969. — С. 62.
  6. Ingber L., Mondescu R. P. Optimization of Trading Physics Models of Markets — IEEE Trans. Neural Networks. 12(4). 2001. P. 776-790.
  7. Ingber L., Rosen B. Genetic Algorithms and Very Fast Simulated Reannealing: A Comparison — Mathematical and Computer Modelling. 16(11). 1992. P. 87-100.
  8. Konstantin Boukreev. Genetic Algorithm and Traveling Salesman Problem [Электронный ресурс]. — Режим доступа: http://www.generation5.org/content/2001/tspapp.asp
  9. Andy Thomas. Solving the Travelling Sales Man Problem using a Genetic Algorithm [Электронный ресурс]. — Режим доступа: http://www.generation5.org/content/2001/ga_tsp.asp
  10. Zach Garner. An Introduction to Genetic Programming [Электронный ресурс]. — Режим доступа: http://www.generation5.org/content/2000/ga00.asp