Artificial Intelligence in Integrated Pest Management

Introduction

Today, modern agriculture is facing great difficulties. Agricultural manufacturing has now become a competitive and globalized industry where farmers and other actors providing input into agriculture have to consider climatic and geographical differences as well as global economic and political factors. In order to ensure nutrition demand of the growing world population, food production needs to be increased from year to year, but the world's arable areas are limited (FAO, 2014; Quarcoo et al., 2014). While the use of agricultural products for bioenergy creates additional pressures on the world’s food requirement, housing as well as transportation problems narrow down the world's arable lands considerably (Boserup, 2017). In recent decades, the changes in global climate has led to an increase in unexpected precipitations, global warming, droughts and the frequency of extreme weather events (Piao et al, 2010). These unfavorable conditions lead to endangering traditional production areas and bringing new risks and uncertainties for world agriculture. Because of climate change, the outbreaks of existing pests are increasing or they become the main pests in areas where they did not cause problems before (Scherm, 2004). An increase in the international exchanges of infested material (seed, plant and soil) is caused by unpredictable pest problems faced by farmers (Dent et al., 1995). In addition to all these global problems, issues related to pests such as pesticide resistance, secondary pest outbreaks and breakdown in host plant resistance, have increased the magnitude of the agricultural production problem (Weller et al., 2014). To cope with these challenges, a continuous and sustainable increase in productivity in all agricultural production areas is needed, while water, energy, pesticides and fertilizers, etc. should be used diligently and efficiently.

Consequently, agriculture needs help to eliminate these problems and uncertainties and requires new solutions for all aspects of agricultural production, from better and predictable product planning to precision agriculture and optimized resource implementation. Nowadays, the most popular and sustainable approach to controlling pests is Integrated Pest Management (IPM) (Flint and Bosh, 1981). However, this approach is quite sophisticated and requires a lot of knowledge, expertise, and observation. In practice, it is necessary to monitor the pest in the field, determine the most sensitive stage of pests, decide on the most appropriate control tactic (pesticide or other alternative methods) and apply it in the best time (Flint and Bosh, 1981). Consequently, IPM requires intensive field observation, trained staff, and data mining. In this context, it has emerged that the use of artificial intelligence (AI) algorithms is a necessity for controlling, tracking, and using these agricultural inputs at the optimal times (Azfar et al., 2015). In this chapter, we will examine the impact and the historical development of Industry 4.0 on modern agriculture and the innovations in IPM in concerning information technologies.