Financial Simulation with AnyLogic

Part 1

The report considered the Base model Add linked to the sales. The simulations focused on the financial reserves with a time base of 10 years. The report further took note of randomness while focusing on the experimental simulations. On the other hand, the presumptions aligned to sensitivity analysis took mote of the profit margin that was expected to be distributed from 5% to 10% with keen focus on uniform distribution. The area of focus is to model a process that would see the simulation seek the verifiable application of the AnyLogic model (Kremers et al. 2013). This would be aligned to agents and processes regarded relevant during the modeling stage. Besides, changes in terms of specific numbers assigned to the parameters within the model created a platform for attracting tools provided by AnyLogic. This would be aligned to optimization procedure that played a focal role in customizing the simulation experiment using external codes (Merkuryeva and Bolshakovs 2010). Perhaps, running a chain of simulations enabled the experimental process to establish a unique setting that suits the parameters in question, or changes in the variables as established in the experiment. Significant changes would therefore pave way for running the simulation for a few times while noting observations within the settings. At the same time, experiments shall check on the replications and computations that determine the memory to be assigned to the simulations.

The sensitive side of the entire simulation process entails the returns, which are largely affected by the optimization procedures determined by AnyLogic simulation model. In most cases, which would apply to this scenario, simulation hubs are likely to manifest in terms of the binary variables that can be updated via the adjusted optimization process. Apparently, most of the developers would set their preference on the simulation checks that can equally be supported by the use of AnyLogic (Karpov et al. 2005). Upon establishing an input that would range from 0.5-1%, which can be assigned as the profit margin, it is more effective to consider an iteration of the simulations in correspondence to the input that needs to be adjusted. Subsequent reruns would end up giving a more harmonized input. With a multiple range of optimizations, it is possible to do analysis in the face of the varying parameters that would be put into consideration (Yang and Zhao 2014). Apparently, the optimization process boosts the probability of exploring more case scenarios and putting them into more comprehensive scenarios noted under different circumstances. A trade-off needs to be established amid the evaluation process as well as enhancing the possible solutions aligned to more specific challenges within an organizational context. Common obstacles would still be encountered in different scenarios. While this may sound like tough times for the developers, this provides a chance to easily lead towards solutions with attempts aligned to either debugging or re-running the code to establish the stability of the code.

The experiment further placed more attention on sensitivity analysis. In this case, the sensitivity analysis looks at how uncertain the outcomes of the re-runs in the mathematical model can be identified across the assignments or the sources given to the system. Across the sensitivity analysis, the Anylogic software has the capacity of establishing the number of codes that would appear in a given range of the iterations that would be assigned to the cores. In equal measures, Randomness could still be established with the help of the code generator with the help of a specific random number. The model could initially be assigned 0.07 (8%) to the profit margin. This would attract a separate sensitivity analysis that would be given the results as shown below.

Part 2

The report reflects on significant modification associated to the discrete event linked to the base model while explaining different case scenarios. The latter narrows down to a new distribution centre while considering the scale of the resources and the type of parameters to be introduced in the system. The significant model can be involved in the modification process that cover a range of parameters such as the ratios assigned to the new center. Extra details assigned to the model include the reporting time of trucks, 2 workers while considering the fact that the parking alone would need only one employee. Figures assigned to the variables include 20% of the goods available which are kept in the freezer while the subsequent 30% can be traced in the chilled storage. The remaining ratio is kept in the well-designed ambient temperature warehouse. While establishing the initial sketches, it was more important to analyze means through which the operations of the warehouse could be optimized. At the same time, the area for the parking space was quite necessary as far as the premier layout is put into consideration. The parameters would work more appropriately in the course of prompting forecasts while engaging the operational needs in the warehouse. The Anylogic simulation model supports efficiency and a wider range of the dynamics related to a number of issues. The essential side of this project can be aligned to designing, optimizing and planning as far as the warehouse operations are put into consideration. Perhaps, the preliminaries of the simulation process require modeling which should reflect the real world case scenario. In the face of engaging the simulation model, it should be noted the primary goal of the optimization process need to be inclined towards the GUI design that needs to reflect the desirable outcome. The absolute goal in this part narrows down to development of a simulation that shall will bolster the operations of the parking yard as well as the warehouse as far as the new distribution centre. The model shall insist on employee efficiency, work efficiency, stock efficiency, yard space efficiency and truck efficiency. For achievement of the work maximization, attaining the efficiency across the parameters demand that the model should set up the most competing or rivalry objectives, which need to be realistic and achievable at the same time via the Pareto optimization. Sampled scenarios give the realistic situations in which the warehouse or the parking yard tends to be full after a few operations.

This implies that the model needs to be evaluated before modeling begins and later, the simulation process. Key functions which can be introduced via the model include the scalable utility of the pallet storage, work and the inflow and outflow of the retail trucks in and outside the distribution space. In the course of developing the new centre, the retailer will put more attention towards the space in the parking yard and whether it allows a rotation of operations or not. The operation simulation model adopted in this context plays the most fundamental role in engaging the logistic system as established in the Anylogic platform. The model carries with it one key objective of constructing the connection across the output and the input parameters, which are also regarded as part of the performance measures. For reasons of deploying the necessary guidelines and logic, the java platform cannot be avoided either and leads developed of classes before working on the entire modeling approach. The pictorial out can equally be enhanced through the appropriate design of the GUI. More focus is put on the flexibility property of the model with the input parameters being introduced to the simulation runs as well as the run time in the face of the efficiency of the warehouse operations. While most of the variables may not be considered in designing the model, the final output may still consider most of them in attaining the best operational mode.

The parking yard optimization would demand that the input values are maintained in case of any adjustments that would be needed by the retailer. The arrivals times of the trucks and the parking time would require the Poisson’s approach while pursuing the probability frequency while relying on the independent events which would occur at given times. Some of the pseudonyms could equally be adopted to produce the prerequisite control charts in the face of determining the discrepancies which would lead to the production of better and more accurate outcomes. Both the outbound and inbound calculations for the warehouse space and the yard would be adopted while determining the operation expenses. The simulation even became more interesting while relying on the significant inputs which could be changed almost instantaneously. Similar scenarios would be encountered while intending to introduce an offloading dock while noting the total expenses. It is also recommended that the distribution centre should keep the communication flowing while taking note of the dynamics and reflecting on the prior model. Most of the inbound yard inputs and outputs will be determined based on the standards noted both in the parking yard and the warehouse while observing the demand conditions. The shopping process and the shopping time are equally important and the central processes need to ensure that their goods flow smoothly and the retail trucks make the most organized moves.

Part 3

In this section, the simulation process considered the demand for observing the dynamics while putting into consideration the possible changes that are likely to manifest in the coming 10 years. A retailer at the new distribution centered, highlighted in the section above, would be concerned to invest close to 3% of the resources in the IP system and research and development framework (Djanatliev and German 2013). The Agent-Based Modeling, denoted as ABM, together with the programming languages played a focal role in determining the design agents. In the course of minimizing the relevant costs, the design model observed significant attributes, behaviors and even agents that would be described by the system under consideration (Wai and Ostroukh 2014). The most observable course involved profiling the model requirements, the relationship across the agents and the necessary mode of interaction. In the course of establishing the requirement, it was necessary to take note of the typology that also applied the elements of the interacting agents. The agent environment is necessary while declaring the conditions around such environment with observations of the transitional states. Three observable elements were necessary while aligning the model to the design and the simulation process. These include the condition or the status, the message that need to be passed across the agents and timeout. Upon introducing a new order, the system prompts the first branch of communication that introduces a query. The query prompts the system and asks whether there are available resources that would help in handling the query. Order lines and order amounts would be produced as the relevant output in the system. However, the system should equally keep an eye on the sales and the costs appended to the order. The customer feedback is equally valued with system checks prompting changes in the design or modification of the system if necessary.

Part 4

State charts are important and play a focal role while designing the market research and establishing the customers’ behaviors. They sometimes reflect on the status of the system. In this context, a series of assumptions were put in place in developing a better system (Zhang et al. 2012). The first presumption is the marketing through the word of mouth and brand attraction would lead to around 30% of the total purchase. 70% which also includes the remaining part would only run for around 5 or a maximum of 10 days if the system does not encounter any problems. However, 25% of the total adopters re expected to make significant attempts while they try to persuade others in making a similar purchase. In case the adopters stop due to the competitive price, then this would imply that no purchase can be realized if there are no Amazon Prime Subscribers. The state chart, therefore, presents circumstances under which a purchase would be made in scenarios that a purchase may fail regardless of the possible efforts. Notably, in the case where a purchase is made, then it must be informed by other factors as well.

References

Merkuryeva, G. and Bolshakovs, V., 2010, March. Vehicle schedule simulation with AnyLogic. In 2010 12th International Conference on Computer Modelling and Simulation (pp. 169-174). IEEE.

Yang, Y., Li, J. and Zhao, Q., 2014. Study on passenger flow simulation in urban subway station based on anylogic. Journal of Software, 9(1), pp.140-146.

Wai, P.A. and Ostroukh, A.V., 2014. Development of simulation model mixed system in the AnyLogic software. International Journal of Advanced Studies, 4(4), pp.48-53.

Karpov, Y.G., Ivanovski, R.I., Voropai, N.I. and Popov, D.B., 2005, June. Hierarchical modeling of electric power system expansion by anylogic simulation software. In 2005 IEEE Russia Power Tech (pp. 1-5). IEEE.

Zhang, Y., Wang, Y. and Wu, L., 2012. Research on demand-driven leagile supply chain operation model: a simulation based on anylogic in system engineering. Systems Engineering Procedia, 3, pp.249-258.

Djanatliev, A. and German, R., 2013, March. Large scale healthcare modeling by hybrid simulation techniques using AnyLogic. In Proceedings of the 6th International ICST Conference on Simulation Tools and Techniques (pp. 248-257). ICST (Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering).

Kremers, E., de Durana, J.G. and Barambones, O., 2013. Multi-agent modeling for the simulation of a simple smart microgrid. Energy Conversion and Management, 75, pp.643-650.