A novel meta-heuristic algorithm, named DHOA, proposed by Brammya et al. in 2019 which is inspired by the hunting behavior of humans toward deer. Even though the activities of the hunters differ, the way of attacking the buck/deer is based on the hunting strategy they develop. Due to the special abilities of deer, it can easily escape from the hunting by the predators. The hunting strategy depends on the movement of two hunters in their best positions, termed as leader and successor. To hunt the deer, the hunters encircle it and move towards based on few strategies. The strategies include the consideration of various parameters, like the wind angle, deer position and so on [1-5]. Accordingly, each hunter updates his position until they reach the buck. Cooperation among the hunters is another important criterion that makes the hunting effective. Finally, they reach the target based on the position of leader, and successor .
Although the main objective of the DHO algorithm is to find the optimal position for the human to hunt the deer, it is necessary to study the behavior of the deer. They have certain characteristics, which make hunting difficult for the predators. One of those characteristics is the visual power, which is five times better than human beings. However, they have trouble seeing red and green colors [7-10]. A deer, also known as a buck, can notice even a slight move and biologists say that a white tailed deer has a peripheral vision that ranges from 250 to 270 degrees. This helps a buck to find the movement of the hunter, but below the horizon. As it is difficult for the buck to pick up the move above the horizon, a hunter can allegedly take movements in a tree top. The sensibility of a white-tailed deer is superior with which it can sense the slightest danger. It can smell sixty times better than humans with its olfactory sensors . On sensing any danger, it alerts other deer by treading heavily and sniffing loudly. The hearing power of the deer is not better than humans, and studies suggest that the possible sensitivity of a deer at moderate frequencies is from 3000 to 8000 Hertz, while the humans can hear from 20 to20 000 Hertz. One of the considerable skills the deer have is its sense of detecting ultra-high-frequency sounds, which a human cannot. The ears of a deer are similar to satellite chips that roll around to catch various sounds .
2.1. Behaviour of White Tail Deer
A sound knowledge of deer behaviour is essential for the efficient management of deer. This guide describes features of deer behaviour which are more or less common to UK deer species (see Species guides for behaviour specific to each), but each species is different, deer are very adaptable, and their behaviour can vary widely with habitat, deer density, and human disturbance.
Deer are usually found in or near to forest/woodland/ scrub and frequently feed on grass or arable land near to cover. Some species are content to live on open areas of heath or moorland. Deer are very adaptable and may be seen in peri-urban or urban situations or other places where there is cover and a food source; they can have marked preferences for different habitats according to deer species, habitat type, and the time of year or weather conditions.
2.1.2. Social Structure
Deer may form herds or act in a more solitary manner according to species, age and sex. Solitary deer tend to be territorial, especially the males. Herding animals are more inclined to form groups; members of a herd may often be closely related. Herds or individuals tend to be “hefted” to an area in which they prefer to live, this tendency is strongest in territorial deer and females of the herding species. The herding deer may have “core areas” sometimes several miles apart, using these at various times of year but rarely being seen on the intervening land. It may sometimes be possible to identify the home range of a herd, group or individual by sightings and the signs that they leave behind (Deer Signs guide). The relationship of young deer with adults varies with species, see Table 2. Where different species live in the same area, they generally live separate lives but there may be passive competition for resources and possibly a degree of aggression.
2.1.3. Response to Disturbance
Deer are relatively shy animals which may give the impression that there are fewer around than there actually are. They are alert to danger and will respond quickly, usually by running away and/or seeking cover but sometimes lying up in or running into wide open areas where they can assess threats. Deer can be easy to stress but recover quickly when the threat is removed. One common response to persistent disturbance is for deer to change their behaviour, e.g. by avoiding busy times/places or becoming nocturnal. This plasticity of behaviour must be taken into account when planning deer management.
Deer are generally quiet but most have specific alarm calls, calls between mother and young and male calls during the rut, some have female calls when seeking a mate or giving birth. Deer tend to use well-trodden paths known as racks to move within cover and to and from feeding places. Using these and other deer signs it is usually possible to build up a picture of how deer use a habitat.
3. OPTIMIZATION BASED ON HUNTING BEHAVIOR OF HUMANS TOWARD DEER
This section deliberates the mathematical model of the DHO algorithm with the steps described below.
3.1. Stages of DHO Algorithm
3.1.1. Population initialization
3.1.2. Parametric initialization
3.1.3. Position propagation
(a) Propagation through a leader’s position
(b) Propagation through position angle
(c) Propagation through the position of the successor
3.1.1. Population Initialization
The primary step of the algorithm is the initialization of the population of hunters which includes the total number of hunters.
3.1.2. Parametric Initialization
Following the population initialization, the wind angle and the deer’s position angle (position angle), which are the important parameters in determining the best positions of the hunters, are also initiated.
3.1.3. Position Propagation
As the position of the optimal space is unknown initially, the algorithm assumes the candidate solution close to the optimum, which is determined based on the fitness function, as the best solution. Here, we consider two solutions, namely leader position which is the first best position of the hunter and successor position which is the position of the succeeding hunter.
(a) Propagation through a Leader’s Position
After defining the best positions, each individual in the population tries to attain the best position and thus, the process of updating the position begins.
(b) Propagation through Position Angle
To enhance the search space, the concept is extended by considering the position angle in the update rule . The angle calculation is essential to determine the position of the hunter such that prey is unaware of the attack and hence, the hunting process will be effective. In this step, angle of visualization of the deer or the prey is formulated.
(c) Propagation through the Position of the Successor
In the exploration phase, the same idea in encircling behavior can be adopted by adjusting the vector L. Since we assume a random search initially, the value of the L considered is less than 1. Therefore, the position update is based on the successor position rather than the first best solution obtained. From the random initialization of solutions, the algorithm updates the position of the search agents at every iteration based on the best solution obtained. When | L |< 1, a search agent is selected randomly, whereas the best solution is chosen when | L | 1to update the position of the agents. Hence, by the adaptive variation of the vector L, the proposed algorithm switches between exploration and exploitation phases .
The position update is done at each iteration until the best position is determined, which is nothing but the stopping criterion, based on the objective function . The Pseudocode of the deer hunting algorithm is given in Table 1. It begins with a random set of solutions. The hunters or the agents adjust their positions following the position update mechanism.Figure3illustrates the Flowchart of the proposed algorithm.
4. PSEUDOCODE OF DHO ALGORITHM
5. FLOWCHART OF DHO ALGORITHM
6. APPLICATION AND ADVANTAGES OF DHO
The DHO algorithm can be applied to various kind of engineering optimization problems such as,
- Pressure vessel design 
- Welded beam design problem 
- Tension or compression spring design 
- Biomedical applications 
- Smart environment [20
- Effectively solve the engineering problems
- Outperforms the existing optimization techniques
- Fine adjustment in parameters does affect the convergence rate of the optimization process
- Improve the classification accuracy
- The DHO algorithm can highly balance the exploration and exploitation phases
- The searching ability of the DHO algorithm for extensively attaining the favorable regions is observed to be high
- The behavior of the search agents is not consistent, and further, it converges properly. Hence, the convergence behavior of DHO algorithm is better than the other state-of-the-art metaheuristic algorithms
 G. Brammya, S. Praveena, N. Ninu Preetha, R. Ramya, B. Rajakumar and D. Binu, “Deer Hunting Optimization Algorithm: A New Nature-Inspired Meta-heuristic Paradigm”, The Computer Journal, 2019.
 Tian, M.W., Yan, S.R., Han, S.Z., Nojavan, S., Jermsittiparsert, K. and Razmjooy, N., 2020. New optimal design for a hybrid solar chimney, solid oxide electrolysis and fuel cell based on improved deer hunting optimization algorithm. Journal of Cleaner Production, 249, p.119414.
 Nassef, A.M., Fathy, A., Sayed, E.T., Abdelkareem, M.A., Rezk, H., Tanveer, W.H. and Olabi, A.G., 2019. Maximizing SOFC performance through optimal parameters identification by modern optimization algorithms. Renewable energy, 138, pp.458-464.
 Tian, M., Yan, S., Han, S., Nojavan, S., Jermsittiparsert, K. and Razmjooy, N. (2020). New optimal design for a hybrid solar chimney, solid oxide electrolysis and fuel cell based on improved deer hunting optimization algorithm. Journal of Cleaner Production, 249, p.119414.
 Yin, Z. and Razmjooy, N., 2020. PEMFC Identification Using Deep Learning Developed BY Improved Deer Hunting Optimization Algorithm. International Journal of Power and Energy Systems, 40(2).
 Onay, M., 2016, May. A New and Fast Optimization Algorithm: Fox Hunting Algorithm (FHA). In 2016 International Conference on Applied Mathematics, Simulation and Modelling. Atlantis Press.
 Mirjalili, S. (2016) Dragonfly algorithm: a new meta-heuristic optimization technique for solving single-objective, discrete, and multi-objective problems. Neural Comput. Appl., 27, 1053–1073.
 Bhushan, S.B. and Reddy, P.C.H. (2018) A hybrid metaheuristic approach for QoS-aware cloud service composition. Int. J. Web Serv. Res., 15,1–20.
 Comi, A., Fotia, L., Messina, F., Pappalardo, G., Rosaci, D. and Sarné, G.M.L., A Reputation-Based Approach to Improve QoS in Cloud Service Composition.IEEE 24th Int. Conf. Enabling Technologies: Infrastructure for Collaborative Enterprises, Larnaca, pp. 108–113, 2015.
 Xin-She, Y. (2014) Chapter 1: Introduction to Algorithms. Nature-Inspired Optimization Algorithms (1st edn), pp. 1–21. Elsevier.
 Fraga, E.S., Salhi, A. and Talbi, E.-G., On the Impact of Representation and Algorithm Selection for Optimisation in Process Design: Motivating a Meta-Heuristic Framework. Recent Developments in Metaheuristics, Springer, Cham, pp. 141–149, September 2017.
 Mirjalili, S., Mirjalili, S. M. and Lewis, A. (2014) Grey wolf optimizer.Adv. Eng. Softw., 69,46–61.
 Mirjalili, S. and Lewis, A. (2016) The whale optimization algorithm. Adv. Eng. Softw., vol. 95,51–67.
 Oftadeh, R. and Mahjoob, M.J. (2009) A New Meta-heuristic Optimization Algorithm: Hunting Search.Proc. 2009 Fifth Int. Conf. Soft Computing, Computing with Words and Perceptions in System Analysis, Decision and Control, Famagusta, pp. 1–5. IEEE.
 Holland, J.H. (1992)Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control and Artificial Intelligence. MIT Press Cambridge, MA, USA, pp. 59–62.
 Kennedy, J. and Eberhart, R. Particle Swarm Optimization. Proc. IEEE Int. Conf. Neural Networks, pp. 1942–8, 1995. IEEE.
 Eberhart, R.C. and Shi, Y. (2001) Particle Swarm Optimization: Developments, Applications and Resources. Proc. Congress on Evolutionary Computation (IEEE Cat. No.01TH8546), Seoul, Vol. 1, pp. 81–86. IEEE.
 Rini, D.P., Shamsuddin, S.M. and Yuhaniz, S.S. (2011) Particle swarm optimization technique, system and challenges. Int. J. Comput. Appl., 14,19–27.
 Yang, X.S. and Deb, S. (2009) Cuckoo Search Via Lévy Flights. World Congress on Nature & Biologically Inspired Computing (NaBIC), Coimbatore, pp. 210–214. IEEE Publications, USA.
 Xin-She, Y. (2009) Firefly algorithms for multimodal optimization. Sapporo, Japan, 26, 169–178.