Last week, our Service Delivery Manager, Pete Johnson, and our Lead Data Scientist, Kathryn McCrindle, were invited to present at the 10th Meetup of the Glasgow Internet of Things group, organised and hosted by Stephen Milne of CENSIS.
Passionate about all types of data, no matter the source or industry, we thought the Meetup was a great opportunity to discuss methods that any company could use to visualise data outputs from smart devices. While the implementation of IoT creates limitless opportunities for data generation and exchange, the true value of new IoT technologies can only be fully realised once data is correctly prepared, cleansed, and analysed. Ensuring proper data management and performing in-depth data analysis are both essential steps that allow businesses to maximise the return on their IoT investment and provide decision makers with trusted, actionable insight from IoT-generated data. As with all data, regardless of type, bad data in can only equal bad data out.
Data Services, Business Intelligence and Analysis can improve any IoT deployment in various ways:
- Improve data quality and confidence from IoT sources
- Integrate and transform multiple data sources to improve later analysis
- Assure security and governance of generated or exchanged data
- Produce dashboards and data visualisations to aid exploration of data for users
- Provide more in-depth reporting
- Create environments that allow self-service data exploration while retaining governance
- Identify trends and patterns within data from IoT sources
- Perform predictive analysis to determine causality between data sets
- Improve decision-making capabilities by providing trusted insights
To illustrate the insights that can be created through data visualisation, Kathryn created a dashboard using one of our partner technologies, Tableau. The dashboard analyses fictitious data created to represent common metrics captured by smart bins in Barcelona. Since 2012, Barcelona has become a perfect example of a city realising the potential of IoT technology by “deploying responsive technologies across urban systems including public transit, parking, street lighting, and waste management. These innovations yielded significant cost savings, improved the quality of life for residents, and made the city a centre for the young IoT industry” (Data-Smart City Solutions).
Kathryn’s dashboard explores how visualisation tools such as Tableau can be used in conjunction with IoT in any setting to further improve efficiencies, cost savings, return-on-investment, business decision making and more. By combining the data generated by smart bins, such as bin location, bin temperature, current fill level and rate of bin filling, we are able to explore the data in deeper and more interesting ways.
Some questions that we could answer from our visualisation include:
- Is there a particular route that needs to be emptied more often to ensure the environmental quality of the surrounding neighbourhood?
- Are there routes that are emptied too often that could be cut back to reduce costs and improve efficiency?
- Has there been any damage to bins as a result of a fire?
- Is there a strong correlation between environmental bin temperature and the rate of bins filling up?
- Is there a way to reorganise city bin collections that is more efficient?
To further analyse the data gathered by our smart bins, we looked at ways we could blend and merge our ‘bin data’ with other data sources to create richer analysis, and provide more information that could drive decision-making. As Barcelona sells a city-wide museum pass, we wanted to explore whether the increase in the number of passes sold correlates with an increase in the rate of our smart bins filling up. This metric could act as an indicator of how busy the city was each day, and form the foundation to determine if increased footfall correlates with the rate of the smart bins filling up.
After integrating our two data sources and creating another visualisation, we noticed that it was likely that a correlation may exist between our two data scources.
By using Tableau, Kathryn was able to perform some statistical analysis on our data to see if our correlation was statistically significant.
As a result of her calculations within Tableau, we discovered that there was indeed a significant correlation between the number of museum passes sold and how quickly our smart bins filled up, at a p<0.05 level. Now, from our existing ‘bin data’, we can make more informed decisions about bin collection by analysing museum pass data, knowing that our decisions are based on trusted, factual insights.
As there is a significant correlation between museum passes sold and bins filling up, we could further investigate our ‘bin data’ by combining it with other indicators of footfall, such as ticket sales or temperature data, to look at further ways to improve rubbish collection routes and times. Not only can visualisation answer questions asked of the data, but it can aid the identification of new questions, making data provide even further business value
More From Our Blog
Welcome to Power BI Premium. In the Final Part of our our Ultimate Power BI Licencing Guide we review Power BI’s Premium licensing option.
As report sharing is essential to realise BI ROI, we’re taking a look at Embedding options for Part 3 of our Ultimate Power BI Licencing Guide.
Find out about the extra precautions we’re taking to ensure both the health & well-being of our employees, while continuing our service provision as normal.