Our Story- Data for Governance Hackathon
Data4Governance Project Documentation
Duration of Hackathon: February 17th-27th, 2020
Team: Reispar Analytics Academy Team
Target Sector: Health
Use Case: Lagos State
Problem Statement: In a densely populated city such as Lagos State, with 20 local governments, there is a major concern for accidents, security and fire outbreaks. There is a need to improve the efficiency of Emergency control services and the response time to help minimize the loss of lives and properties due to these situations. Hence the incentive for this project.
Aim of the Project: To Improve Emergency Response Time in Lagos State, Nigeria.
Objectives:
- To identify existing fire emergency units
2. To identify towns and local government areas that were more sensitive to fire outbreaks
3. To identify LASEMA response units within these local government areas.
4. To map out the proximity of LASEMA response units to these susceptible local government areas.
5. To identify possible areas for sighting these fire service stations
1 Team, 5 Tools, 5 Unique Skills
17th February 2020, the hackathon commenced at CCHUB, Sabo, Lagos State, Nigeria, with 20 selected teams. The teams were introduced to, provided links and access to databases on various sectors such as Health, Transportation, Education, Agriculture, Urban planning and Citizen Accountability. Each team was required to brainstorm and choose a sector to work with.
Our team chose to work with and come up with a solution in the “Health sector”. With a combination of skilled individuals across Data mining, Data wrangling, Data Engineering, Data visualization, Design, and Geospatial analysis, we embarked on the 8-Days “Data4Governace visualization” hackathon powered by CCHub in partnership with the World Bank to accelerate the utilization of data for good governance and improved Public service delivery.
Use Case :
In recent years, there has been a significant increase in the number of lives and properties lost to fire outbreaks in Nigeria. Few examples of states that have been greatly affected by this are:
- Abia State: Pipeline explosion in Aba in October 2010 which recorded about 150 deaths
- Lagos State: Fire explosion in Balogun market on January 29th, 2020 and Apongbo fire explosion on February 16th, 2020.
- Bauchi State: 2 lives lost, 147 lives saved, N948,025,000 properties destroyed in a year (between June 2018 and June 2019) as a result of home and industrial fire outbreaks.
- Lagos State: a Truck explosion on Lagos-Ibadan expressway, 25th February 2020
Response Agencies — Lagos State Emergency Management Agency(LASEMA) & NEMA(National Emergency Management Agency)
The Lagos State Emergency Management Agency is an initiative from the Lagos State Government that coordinates the Lagos Response Unit. It is charged to provide an adequate and prompt response as well as sustaining intervention in all forms of emergency/disaster situations in the State within the territorial boundary called “Lagos”.
LASEMA Response Unit (LRU): LASEMA Response Unit is an Emergency care service that is tasked with the responsibility of preventing and mitigating emergency disaster such as fire outbreaks, road traffic injury, curbing the spate of collapsed buildings in Lagos State, amongst others; using a toll-free call code of 112.
With the increase in urbanization in Lagos State, there has been a rise in fire outbreaks occurring in homes and public places. Most of the incidents have witnessed slow response time from the emergency control centers. The number of lives and properties lost to fire outbreak has hence heightened the need for improving the efficiency of these Emergency response stations within Lagos State, Nigeria.
Data Sources
With the instructions and guidelines given, we optimized the use of the data found in the KOBO TOOLBOX platform and GRID3 website. The Kobo Toolbox is a suite of free and open-source data across transport, security, agriculture, health, climate change, etc to gather and analyze crucial information. Our major data source was the GRID3 website owned and managed by the Federal Government of Nigeria. The data collected were on hamlets, markets, fire and ambulance services, etc.
Collect → Clean → Visualize
With the data collected from Grid3, we identified relevant data required for our target sector and use-case, then modified the data using DataScience tools such Python and its rich libraries one of which is the “Geopandas Libary”, thereby ensuring the data is credible and fit for the intended purpose and stored as a CSV file for visualization purposes.
For visualization, we utilized Data Analytic tools such as “Microsoft PowerBI” and “ArcGis”.
View Visualization on Power Bi Here:
Data Insights:
The following data was used to draw insights, with a major focus on:
https://docs.google.com/spreadsheets/d/1ifr8BTyud0HO2D64jWWBQQbpNnC8HTEPE9vK-UCD73Q/edit?usp=sharing
- Response Elements in Lagos (Fire Stations):
- Towns in Lagos State
- Local Government Areas in Lagos State
- Markets in Lagos State
Methodology:
- Using historical data (past occurrences ) to gather highly emergency scenarios and areas grouped by their Local Government Area (LGAs).
- Map-out case study areas based as assessed from their LGAs
- Research and list out Fire Service Stations based on a field research visit.
- Network Analysis to show the shortest route between Fire Service Stations to the Towns (under Wards) wherein they are located at.
- Based on the result from Step 4, map-out possible areas for sighting Fire Service Stations in a Ward.
“No doubt fire can occur anywhere, however, though our network analysis considering Fire Station-Town bases, we can infer that Alimosho is the LGA most affected and will require more sightings of Fire Stations”
Expected Outcomes:
- Using Power BI and ARCGIS to show less focused areas and the relationship between Towns and Fire Service Stations.
2. Hotspot Cluster Map/ Heat Map (market and Fire station services): The heat map for market data focuses on high-risk areas using market places in Lagos as a case study. The baseline for this data was the number of market days, as expected higher chances of fire. Also, a heat map for locations of Fire service stations was done.
3. Using Power BI report to show less concentrated areas (with fewer or now Fire Service stations/ response centers). Automatically, this will increase the number of response time from the Fire service stations.
4. Buffering & Analysing Fire Service Areas-Towns: We used a buffer feature on ARCGIS with a coverage range of 4000metres to create a spatial visual of Fire Stations in Lagos and proximity to Towns. The results made us aware of areas with more clusters of Fire Service points (high concentrated) to Towns. That is, some areas in Lagos have fewer or no Fire Stations close by.
5. Network Analysis and Table of Response Time: The network analysis shows the time it will take from a Fire Service Stations to a Town in a less concentrated area (areas with fewer Fire service points).
This data and project enable us to proffer solutions in improving response time in Lagos.
Recommendations:
- Presence of response units in largely dense areas such as Alimosho and Ajeromi-Ifelodun
- Citing fire service stations in less clustered areas within a local government area to facilitate faster response time
- Readily available fire hydrants in areas prone to fire outbreaks.
Potential to scale
This solution has widespread applicability across the country. We believe the government is very interested in protecting lives and preventing loss of property wherever applicable to the optimum. Hence, there is a need to reduce significantly the losses incurred from fire incidents. The key to this solution is information on prone areas and how response times can be improved. Our project has been able to not only provide insights on these but also better inform how they can be solved. This solution is also applicable across multiple cities in the country.
LINKS TO GEOSPATIAL VISUALIZATION ON ArcGIS:
Route map for the less clustered areas → https://arcg.is/KSq4
Heat map for the fire service → https://arcg.is/1yr4GH
A buffer of fire stations → https://arcg.is/10Wjqa
References:
https://ijerat.com/uploads/2/3693_pdf.pdf
https://www.vanguardngr.com/2018/10/how-pipeline-fire-roasted-over-150-in-aba/
https://www.concise.ng/2020/02/16/fire-engulfs-popular-apongbon-market-in-lagos/
https://dailypost.ng/2019/08/14/statistics-fire-incidents-bauchi-state-emerge/
https://lasema.lagosstate.gov.ng/
https://www.channelstv.com/2020/01/29/breaking-fire-guts-plaza-in-balogun-market/
https://en.wikipedia.org/wiki/List_of_Lagos_State_local_government_areas_by_population
Thanks for reading. Kindly give a clap and drop feedbacks
