Decision Making Using System Modeling: Smart City Policy Report

This report presents a comprehensive smart city policy focused on traffic management and crime reduction, utilizing system modeling techniques and the EMA Workbench. The policy proposes the strategic placement of traffic lights, integrated with smart systems and CCTV cameras, to optimize traffic flow and provide real-time data for efficient decision-making. The report also details the implementation of CCTV cameras equipped with alarm systems and facial recognition capabilities to detect and prevent crime. Furthermore, the study explains the application of the EMA Workbench, Vensim, and the development of static and adaptive policies to simulate and analyze the impact of the proposed interventions. The analysis includes defining uncertainties, outcomes, and the creation of Vensim code to evaluate the effectiveness of different policy scenarios, contributing to improved infrastructure, better governance, and enhanced safety within a smart city framework.

Running Head: DECISION MAKING USING SYSTEM MODELING 1
Decision-Making Using System Modeling
Your name
School-affiliated

DECISION MAKING USING SYSTEM MODELING 2
Introduction
Policy-making is an important activity for any organization, body or society. It ensures
structures and ways of doing things are improved, i.e. improved infrastructure, better
governance, safety, etc. This paper seeks to formulate a policy for a smart city and explain how
to create the model using the open-source EMA Workbench. This software integrates multi-
model, multi-method and multi-policy simulations with visualization, data management, and
analysis (Marijn Janssen, 2015).
a) Traffic light placement and Surveillance Camera Coverage
Traffic management is a key element in every city. It determines how organized the vehicle
movement is, ensures easy flow and maneuverability of traffic and makes sure everyone gets
going to their destinations safely and in good time. A lot of cities have problems with traffic
management and caused a day to day problem to motorists and pedestrians alike. It also puts a
drag on the social-economical activities of the city. This is why this policy is important, to be
implemented in a city that seeks to be smart. This policy implies that there should be enough
traffic lights placed on all junctions, intersections, and roundabouts. The traffic lights should be
fully functional and have smart systems that are able to communicate with android devices
belonging to licensed drivers. The drivers with this app on their phones should be able to alter
the traffic lights in emergency situations in case of very little or no traffic. The traffic lights
should be integrated with the city surveillance to determine when to allow altering of the traffic
lights to allow a motorist with an emergency probably an ambulance to be allowed to pass
through without having to wait for the lights.

DECISION MAKING USING SYSTEM MODELING 3
The city should be fitted with CCTV cameras on every traffic light placement, street corner,
and alley. The CCTV cameras on the traffic lights will help detect the flow of traffic and
intelligently be able to show the red light only when necessary. That is, when one part of an
intersection, junction or roundabout has no vehicles, the traffic lights should avoid showing a red
light on the other part of the intersection, junction or roundabout and allow traffic to flow
continuously. Also, the traffic lights should be able to allow more time on the more congested
part of the intersection, junction or roundabout. The CCTV cameras on other parts of the city
should be fitted with alarm systems and sensors that will intelligently detect crime when it
happens and automatically raise an alarm and alert the police. The cameras will make use of
facial recognition and use artificial intelligence to make a judgment on actions to be able to grade
them as crimes. The cameras should also be able to identify a suspect if he/she appears in view
using the integrated databases from the police department. The CCTV cameras will also help the
government collect information on its citizens for better services to them. The ICT technologies
can be used by IP-CCTV system designer to create a network of CCTV that can operate across
the corporate networks and be able to span across the city and even connect to CCTV networks
in other cities.
Using the EMA Workbench
The EMA workbench can be connected to Vensim directly. We will use in Vensim in this model,
hence we will use the base class Vensim.ModelS.
The outcome desired:
 state of traffic movement.
 level of crime.

DECISION MAKING USING SYSTEM MODELING 4
The outcomes are added to self.outcomes by use of the TimeSeriesOutcome class. This table is
derived from Pruyt & Hamarat (2010) showing the uncertainties and their bounds. They will be
added to self.uncertainties as ParameterUncertainty instances.
Parameter Limit Lower Limit Upper
traffic congestion 0.3 0.8
crime rate 0.3 0.7
Policies are developed using the system understanding of analytics.
 static policy
 adaptive policy
 running the policies
The policies are specified in order to be able to run the models containing the policies then
comparing the outcomes with a no policy case. The policies are implemented in separate vensim
files. One may use any keyword arguments for the policies and if it is a match for an attribute on
the object of the model it is updated. The attribute for each policy is updated during policies
executing. The policies are passed to perform_experiment() as a keyword parameter added. This
is the resulting Vensim Code:
import numpy as np
from workbenchEMA import (TimeSeriesOutcome, real_parameter, EMALogging,
perform_experiments, scalar_outcome)
from workbenchEMA.connectors.vensim import VensimModel
from workbenchEMA.frameworkEM.parameters import Policy
if _name_== '_main_':
EMALogging.log_to_stderr(EMALogging.INFO)
modelSC = VensimModel("smartCity", wd=r'./models/sm,
del_file=r'SCvensimV1basecase.vpm')

DECISION MAKING USING SYSTEM MODELING 5
# outcomes
modelSC.outcomes = [TimeSeriesOutcome(state of traffic movement),
TimeSeriesOutcome('level of crime'),
ScalarOutcome('max degradement fraction', variable_name=degradement fraction
R1', function=np.max)]
# Plain Parametric Uncertainties
modelSC.uncertainties = [RealParameter('traffic congestion', 0.3, 0.8),
RealParameter('crime rate', 0.3, 0.7)]
# add policies
policies = [Policy('no policy', modelFile=r'SCTrafficCongestion.vpm'),
Policy('static policy', modelFile=r'SCTrafficCongestion.vpm'),
Policy('adaptive policy', modelFile=r'SCCrimeRate.vpm')]
results = perform_experiments(modelSC, 1000, policies=policies)

DECISION MAKING USING SYSTEM MODELING 6
References
Kwakkel, J. (2011-2018). Exploratory Modelling and Analysis (EMA) Workbench. Retrieved
from EMA Workbench documentation: https://emaworkbench.readthedocs.io/en/latest/
Marijn Janssen, M. A. (2015, June 3). Policy Practice and Digital Science: Integrating Complex
Systems, Social Simulation and Public Administration in Policy Research. Springer.