Logistics, particularly over the past few years, has been a decisive factor in the growth and development of countless organizations. New manufacturing technologies have enabled companies to lower their costs like never before; but eventually a time comes where cost reduction reaches its limit – and where logistics can mark the difference between business success and failure.
Unfortunately, uncertainty is considered to be the biggest problem associated with supply chain management, according to some experts. Variability and risk are the main difficulties in any logistics network, and only adequate planning, along with strong communication can deal with those difficulties. (Mason-Jones and Towill 2000)
According to Anish Jain, head of the Institute of Business Forecasting, 7 out of 10 companies in Mexico make critical mistakes on their business forecasts, affecting their productivity and prime management of their supply chains, putting their competitiveness and profitability at serious risk.
In the face of globalization and competition around the world, but also because of other external factors outside of a company’s control (political, economic, social, technological, environmental, and legal) it is important for businesses to adequately forecast the demand of their products or services, in order to have operational and sales plan that will allow them to know when and how much to buy or to produce, in what quantities, and how to distribute. (Romero and Romero 2008)
Then again, the purpose of each forecast is to reduce the uncertainty of natural variability through the anticipation of future events, and to generate valuable information to support the decision-making process. There is no doubt about how challenging demand forecasting is, but there are different methods and techniques depending on each company’s needs and data availability.
The following are brief descriptions of some of the tools that could be handy, ranging from those based merely on intuition, to the most sophisticated deterministic models:
Subjective or qualitative methods – based on judgment and experience and utilized when the situation is imprecise and little data is available.
- Educated guess: A personal judgment based solely on experience and intuition. It usually works for the short-run, and when the consequences are not so expensive.
- Jury of executive method: This is based on the idea that the intuition of several people is presumed to be superior to that of a single person. It is about discovering an intermediate point of view among various specialists or members of different departments within the company to develop the sales forecast.
- Delphi method: The results of this technique are determined by a consensus of the responses given by experts, collected through questionnaires answered anonymously, taking into consideration that the participants may not be directly involved with the company’s activities. (Romero and Romero 2008)(Corporación Financiera Internacional 2014)
Objective or quantitative methods – based on the analysis of historical data and utilized when the situation is relatively stable.
- Causal models: They use regression analysis to relate those independent variables or factors, such as time, price or publicity, to a dependent variable (demand).
- Time series: A time series is a set of evenly separated numerical data obtained through the observation of responses within regular time intervals. Time series are exclusively based on past data about the phenomenon intended to be studied or forecasted. (Romero and Romero 2008)(Corporación Financiera Internacional 2014)
It is important to mention that aggregate forecasts tend to be more precise and that the wider a forecast horizon is, the lower its precision will be. In addition, companies have to remember that although a forecast may not be 100% accurate, it is essential to make an effort and understand that poor planning can bring negative consequences such as an inability to satisfy customers’ needs, excess inventory, and an increase in logistical costs, among other problems.
Corporación Financiera Internacional. 2014. “Pronóstico de La Demanda.” Accessed March 24. http://mexico.smetoolkit.org/mexico/es/content/es/416/Pron%c3%b3stico-de-la-demanda.
Mason-Jones, R., and D. R. Towill. 2000. “Coping with Uncertainty: Reducing ‘Bullwhip’ Behaviour in Global Supply Chains.” Supply Chain Forum. An International Journal, 40–45.
Romero, O., and S. Romero. 2008. “Pronóstico de La Demanda.” Instituto Tecnológico Autónomo de México. July 11. http://allman.rhon.itam.mx/~oromero/Notas3_Pronostico_e_Inventarios_Diplomado_Plan_y_Dir_Ope.pdf.