Smartening the movement path of municipal garbage trucks using genetic algorithm with emphasis on economic-environmental indicators

The collection is one of the most important steps in waste management, accounting for 60% of total costs. Therefore, a little improvement in collection operations can have a signifi cant impact on total cost savings. On the other hand, the traffi c of heavy vehicles collecting waste causes the air pollution spread and the passages pavement damage in case of excessive loading. Therefore, the issue of vehicle route determining to achieve this goal is very important. This study simulated the routing process of garbage trucks using random routing problems and genetic algorithms. The simulation results showed that the genetic algorithm converges to the optimal response in the 2069th generation and according to the convergence graph, in the 1000th generation onwards, the slope of the graph decreases. On the other hand, the amount of cost function is reduced from 11775.4909 to 1589.6028 by optimizing mentioned model, and the performance result has led to the emergence of the shortest possible path. With the help of the algorithm, all the management parameters of sustainable development, including reducing air pollution, reducing street pavement destruction, and energy (fuel) consumption are achieved. Finally, by integrating ArcGIS software, the output of the algorithm was matched to the map.


Introduction
The need (demand) to change is the most important natural need of a man in history, promoted along with intellectual and cultural development and social and scientifi c promotion.
This need to change also increased and world civilization owes a human need for change [1]. In fact, this natural human tendency to change and evolve is the source of social, cultural, and political changes. And this sense of perfectionism shows in public areas and topics such as issues related to municipal waste management [2]. The main steps of the waste management process include 6 steps: 1) waste generation, 2) handling, separation, storage, and processing at the source, 3) Collection, 4) separation, processing, the transformation of waste, 5) Transport, 6) Disposal [3]. According to the inevitability of waste generation or waste by humans, the attention to control and regulate (organizing) is a national duty.
In this regard, at the fi rst place should have attention to how to collect large amounts of dust and at a later stage to consider the issue of recycling. Waste collection is actually an attempt to reduce entropy (disorder). The various components of municipal waste are scattered at great distances, and the role of the collector is to collect these materials in one place [4].
In Iran (like many other parts of the world) waste collection systems are person/truck systems. With very few exceptions, municipal solid waste collection is collected in all parts of the world by men traveling around the city by truck. The collected waste is transported to the landfi ll. This can be a fi nal disposal https://www.peertechzpublications.com/journals/annals-of-environmental-science-and-toxicology Citation: Ghadami  site or a temporary station where waste is dumped into larger containers, trailers, ships, or railroad cars at closer distances and then transported to a landfi ll, compost site, or recycling plant [4].
A large part of the cost of municipal waste collection is related to the collection phase (about 60%). Therefore, efforts to optimize this process and improve it began many years ago [5].

Material and methods
Determining the route of a garbage truck in its assigned area is called micro-routing and determining the route of the device for transporting garbage to the disposal site is called macro routing. The methods, limitations, and conditions of these two issues are different. The purpose of routing is to fi nd the shortest path and in fact the path with zero repetition [4].
It can be said that all the proposed models are based on Euler's Tour and the Chinese Postman Problem Qwan. Mathematical models such as time window and meta-heuristic processes, geographical models, and the use of various algorithms such as neural networks.
Of course, it should be noted that in order to adapt the output of the model to reality and increase its effi ciency, in some cases, sensory routing (based on logic and human experience) is used or thus modifi cations are made to the path provided by the model. However, the present study is considering the criteria of distance (which can be considered equivalent to fuel consumption and of course the volume of air pollution caused by it), type of device (affecting fuel consumption and pollution), and pavement of the surveyed areas, and is trying to fi nd a route with minimal fuel consumption, air pollution production, and pavement damage while serving all parts of the zone. This study tries to use a genetic algorithm to hit this goal.

Genetic algorithm
Genetic algorithm is one of the random search algorithms whose idea is derived from nature and has many applications in solving optimization problems. In nature, better generations emerge from a combination of better chromosomes. In the meantime, sometimes there are mutations in the chromosomes that may improve the next generation. Genetic algorithms also use this idea to solve problems. In fact, two things happen to chromosomes at the same time. The fi rst event is a Mutation; This means that some genes are randomly changed. Another thing that happens, and of course this happens much more often than "mutation", is that two chromosomes stick together in length and exchange some pieces between two chromosomes which is known as the Crossover. In fact, genetic algorithms use Darwin's principles of natural selection to fi nd the optimal formula for predicting or matching patterns.
Genetic algorithms are often a good choice for regressionbased prediction techniques.

Modelling and algorithm defi nitions
But we must note that each pavement has a certain amount of tolerable resistance, which W Ti must be less than that. So according to Equation (2) we have: The cost function (2)

Cos 2 min( )
Cost function (3) The route of trucks at specifi c distances according to the instructions in Figure 1 is shown in the diagrams.    The general operation and load distribution of fl exible pavements are shown in Figure 2. According to the fi gure, the vehicle load is applied to the pavement surface by the wheel at the almost circular contact surface. The thickness of pavement layers and their resistance quality are selected so that the wheel load is distributed and spread as much as possible, to reduce the maximum stress intensity on the sub-grade to a lower pressure than the bed soil can tolerate the allowable deformation [12].
Every year, a signifi cant budget is spent on the construction and maintenance of road networks, and therefore it is necessary to be careful in the use and protection of expensive asphalt layers. These loads reduced pavement life and the road was destroyed before the end of expected service life.
Early damage caused additional and unforeseen costs and inconsistencies with the planned schedule and budgets allocated for maintenance [11]. This creates many problems in macro-management issues. In examining the sensitivity of pavement to different types of loads in case of different loads, many regulations tables were examined. The most important ones in this study are Tables 1 to 5.

Results and discussion
According to the framework defi ned in Section 2, the view

Conclusion
The waste generation that inevitably occurs by all city dwellers; On the one hand, and the rapid development of