- 1 Work Package 5: Prediction models
- 1.1 General information
- 1.2 Deliverables
- 1.2.1 D5.1 - Fisheries and Aquaculture Competitiveness Index
- 1.2.2 D5.2 - “Boom-and-bust” model
- 1.2.3 D5.3 - Strategic positioning model
- 1.2.4 D5.4 - Demand analysis model
- 1.2.5 D5.5 - Innovation and price analysis
- 1.2.6 D5.6 - Scientific review paper “Early warning signs for “boom and bust” cycles”
- 1.3 Links & Resources
Work Package 5: Prediction models
DEVELOPMENT OF SIMULATION AND PREDICTION TOOLS
In WP5 simulation models are developed to analyse how changes in supply and demand affect production planning, economic performance, supply chain relationships, value added, potential product success, market trends and developments and thus competitiveness as measured by the updated FACI developed in WP1. Simulations of “boom and bust” cycles will be carried out and common traits highlighted that facilitate the development of prediction models. Building on work undertaken in WP4, a set of product success indicators will be established designed to indicate probability of successful launches on a targeted seafood market. The outcome of this WP will be simulation/forecasting models for analysing changes in competitiveness, prediction of instability of demand and supply including that of warning signs for “boom and bust” cycles and for indication of potential for product innovation success (SO5).
The concept of competitiveness can be traced back to early writing on economics in the 17th and 18th centuries, but has become ever more urgent in the last decades with rapid improvements in transport and communication and a higher level of globalisation. Although competitiveness may be measured by single indicators, such as productivity of labour, a deeper understanding of the competitive standing of firms and countries can be gained by employing multi-dimensional measurements. Currently, the Global Competitiveness Index (GCI), developed and compiled by the World Economic Forum, is probably the most comprehensive index of its kind. The choice of methods depends on a variety of factors, including the perceived need for complexity, data availability and how the results are to be used. (Click title to read more)
This deliverable discusses the methodology to develop simulation and prediction models to be used to predict price behaviour and to be integrated in the GRA of the PrimeDSS and PrimeDSF.
The statistical model used for the price prediction is based on Robust Monitoring of Time-series approach. Time-series often contain outliers and level shifts or structural changes, and these unexpected events are of the utmost importance in the forecasting of prices. The presence of such unusual events can easily mislead conventional time-series analysis and yield erroneous conclusions. The model provides a unified framework for detecting outliers and level shifts in short time-series that may have a seasonal pattern. The methodology was developed to detect potential fraud cases in time-series of imports into the European Union, and we have borrowed the methodology as it is particularly well suited to the type of data and phenomena this deliverable deals with (Barabesi et al, 2016; Fried et al., 2012; Galeano and Pena, 2013; Riani et al., 2012), Rousseeuw and van Driessen, 2006; Salini et al., 2015; FSDA).
The deliverable will be a model based on analysing strategically where in the value-chain companies choose to position themselves and how changes in their environment can affect this choice.
Some consumers expect clean and clear labels, transparency from manufacturers and highest safety while others value great taste, sensory appeal and premium quality. Others are relying on branded products and exhibit loyalty, again others may shop in non-traditional channels for food and purchase based on price. In order to address such consumer diversity and to succeed in a highly competitive marketplace, firms must understand differences in consumer preferences and behaviour in order to address them efficiently. New products (and existing ones) must be connected to consumers’ wants and expectations in order to be placed and marketed strategically and successfully. Many companies struggle with innovation and new product commercialization as is evident in failure rates of new food/drinks products as high as 70-80 %. (Click title to read more)
Building on the consumer analysis conducted in WP4 and in particular in Task 4.4, this deliverable will provide a deeper analysis of the willingness-to-pay (WTP) of consumers, and consequently the price that producers may charge in different markets. This will be done by analysing in depth the relationship between the various products’ attributes and WTP. This deliverable also provides instructions, already circulated to the Syntesa partners, on how to implement these models into the PrimeDSS in WP6.
Scientific review paper “Early warning signs for “boom and bust” cycles”