Excursion Groups Flow Forecasting Based on Modified Random Forest
MAGAZINE №4 (99) August 2020
AUTHORS ASLAKHANOV A.R., PAVLOVA E.V.
CATEGORY Information technologies in logistics and SCM Logistics service management Optimization and mathematical modelling Simulation modelling
ABSTRACT
In recent years, both in Russia and worldwide, there has been an annual increase in the number of museum visitors with the most popular exhibitions being visited by millions of people. In 2020, in the context of quarantine measures caused by the COVID-19 pandemic, the issue of managing museum visitors flows has become especially acute. If earlier museums throughput was limited by the maximum duration of possible evacuation from a museum building, exhibition space and the number of employees working with the visitors, in 2020, due to the observance of sanitary and epidemiological rules, the throughput of museums was further reduced. This determines the relevance of analytical solutions for museums as in order to manage visitor flows and adapt services to the high demand, it is necessary to have an effective forecasting model that takes into account the determinism of demand by a number of factors. The purpose of this paper is to develop a forecasting model for the number of excursion groups in specification museum-day-hour. A modification of random forest with the inclusion of more than 450 independent variables in the model is proposed as a forecasting method. The modification of the model relies on changing the mechanism for combining forecasts of trees in the forest in such a way that the weight of the tree in the model is inversely proportional to the measurement error of this tree. The proposed model is tested on the basis of data on more than 20,000 excursion groups of the State Russian Museum for the period 2018-2020. The proposed model showed high accuracy (36.6% WAPE and 0.5% BIAS).
Keywords: forecasting machine learning random forest combining forecasts Python forecast museum Service service quality
Between a Rock and Hard Place: Logistics Peculiarity Throughout Promotions
MAGAZINE №2 (91) April 2019
AUTHOR BORISOVSKAYA M.A. - Sr Demand Planner, Business Intelligence/Commercial Department, The Coca-Cola HBC EUROASIA (Moscow, Russia)
CATEGORY Analytics and reviews Logistic integration and coordination Supply chain planning
ABSTRACT
Promotions for fast moving consumer goods (FMCG), especially price promotions are among the most powerful tools to increase sales. The sales volume at the promotional price can be ten times higher than at the regular one. The most stable promotional programs today are offered by National Key Accounts (NKA’s).
Herewith, they put forward strict cooperation conditions which are often difficult to fulfill even for transnational FMCG corporations represented in Russia. This review article reveals the trends in the development of relations between the Russian NKA’s and the largest FMCG companies. The role and features of logistics of both contractors in the conditions of promotions are defined, the tasks and problems faced by logistics of both contractors are consolidated. The methodology of the study of cross-functional and inter-organizational coordination is based on system analysis, classification, analogy and generalization, forecasting and interviewing. Top managers of the interacting departments of FMCG companies and NKA’s were interviewed for difficulties in planning and implementing promotions in oral and written forms. As a result, a table of the main points of interaction and possible conflicts among departments within and among FMCG companies and NKA’s has been developed.
Keywords:
Application of Cluster Analysis in Inventory and Assortment Management
MAGAZINE №6(89) December 2018
AUTHORS
MILOV A.S. - Bauman Moscow State Technical University (Moscow, Russia)
CATEGORY Analysis in logistics and SCM Inventory management Retail companies’ corporate logistics
ABSTRACT
The application of cluster analysis requires serious advance preparation of business processes of profit companies' trade networks. But the benefits appearing after its implementation are unique.
- There is the only practical opportunity to manage the assortment and improve the forecasting accuracy with the minimal availability of sales statistics on product stock keeping units or its total absence. This is especially true for both collector assortment with its annual rotation and new positions of traditional and seasonal assortment.
- Subject to prompt performance of management and forecasting tasks it is possible to run down to the level of maximum detailing for each point of trade network realization, (and (or) each customer) and each commodity stock keeping unit. At this level traditional methods of assortment management do not work because of the high speed of decision making.
- It becomes possible to automate the proposed processes in partnership with experts and neural network algorithm, for each realization place, which is an important task for trade networks for the purposes of «PSM technology». Taking into consideration a high level of processes detailing and traffic it is an extremely important achievement.
It became possible to create a unique assortment matrix for each point of purchase of goods (and (or) for each customer) and manage it online. This benefit allows to satisfy customers' demands better and, as a result, increase the financial flow of the company massively. Additional options of assortment management at the operational level have been received as a useful bonus
Keywords:
Application of PSM technology for planning and management of assortment in B2B and B2C companies
MAGAZINE №6 (83) December 2017
AUTHORS MILOV S.N.
CATEGORY Analysis in logistics and SCM Inventory management Logistics service management Retail companies’ corporate logistics
ABSTRACT
Nowadays Russian trading companies have to carry out their product portfolio policies in conditions of tough competition, presence of numerous goods vendors in the market and constant product portfolio renewal. These factors significantly complicate the process of sales forecasting. On the other hand, increasing the accuracy of planning through assortment management at the operations level allows you to use additional features to increase profitability. Several articles will describe practical application of the author»s concept entitled “Product Status Management” in contemporary B2B and B2C companies in order to improve the quality of planning. It took more than 15 years to gather information and formalize this concept in the course of practical work. Initially, the author worked on the issue of elaboration of mathematical forecasting methods by stock keeping units (SKUs). However, in practice he became convinced that the way of solving this problem largely lies in the sphere of administration, management of the product portfolio and the Roles involved in the process rather than in the plan of improving the mathematical methods of forecasting. The proposed «Product Status Management» concept is a logical development of «Product Life Cycle» theory (Theodore Levitt) as well as ECR, S&OP and CPFR concepts. The concept of «Product status management» gives additional opportunities for profitability management and profit increase. It significantly simplifies the subsequent automatization of SCM processes. This article will introduce you to the basic elements of this concept relevant to the planning process.
Keywords:
