Heterogeneous Sensing Data Analysis for Commercial Waste Collection
Abstract
:1. Introduction
- Business data flows composed on the one hand of data imported from PoL’s Enterprise Resource Planning (ERP, e.g., customer inventory) and on the other hand of needs and constraints reported by PoL’s customers and defined according to their socio-professional categories.
- Waste bin filling levels, regularly transmitted by ultrasonic sensors to be deployed as part of the SWAM project.
- Collecting GPS information (trucks fleet) for real time monitoring.
- Deploying ultrasonic filling detection sensors in several bins to capture real life data.
- Proving a comprehensive analysis of the current state of the art in the area of SWC and a technology evaluation to determine the basis for such an approach.
- Proposing and validating a data management system, collecting relevant SWC-related data sources while ensuring privacy and security aspects. Such data are often unavailable. The collaboration with PoL enabled the access to a large dataset with a high time resolution, collected over several years in Luxembourg (2015–2019).
2. Related Work
- Sensor-based waste collection (e.g., dynamic adaptation of trucks routes, and optimal planning and scheduling of waste collection trucks).
- Waste transport-based location intelligence (e.g., route optimization based on the waste type).
- Recycling solutions.
3. Use-Case
3.1. Scenario
3.2. Impacts
3.3. Architecture and Parameters of the System
- Data Collection: it includes different datasets that are captured from several resources. This involves the historical data from the SWC company, namely ERP data while the sensor data are collected with ultrasonic sensors deployed around Luxembourg. Finally, the vehicle tracking information would be used to identify its location in nearly real-time. A relational database is used to store all the collected data (SQL server). Details of each dataset are described in Section 4.
- Data management service: managing varying data resources becomes quickly complex. After the completion of the collection phase, a filtering process would be conducted out to explore the accuracy of the generated information. Predictive models will handle missing fill-level data. Regression mechanisms will enable prediction of the filling level at least 48 hours in advance.
- Optimization service: it is mainly divided into three parts:
- Customer selection: the relevant customers are selected for an inclusion into a waste collection, by considering a set of objectives and constraints.
- Dynamic Scheduling: this module is designed to offer optimal planning and waste collection scheduling (i.e., waste collection would be executed only if it is needed).
- Dynamic Routing: the selected customers are integrated into a fleet management optimization engine that generates a dynamic adaptation of trucks’ routes and selects the optimal path.
- Decision Support System: this interface involves three main modules:
- Dynamic Pickup Points: this module aims to generate automatic pickup points that should be considered by the driver.
- Shortest Path: this module is responsible on recommending an optimal path based on the current vehicle location and the ordered list of clients to be served.
- Activity Recognition: unlike most of the prior studies that have not considered the driver behavior, the proposed system aims to use smartphones and use the embedded motion sensor in order to capture the driver’s activities. Detecting what a driver is doing at a specific point of time would enable the creation of dynamic profiling for each driver independently. For instance, calculating the service time of each collection per site could give an indication of the actual number of processed bins.
4. Datasets
4.1. ERP: DIVALTO
- Client
- ClientName/ClientNameId refers to the client’s name in a human readable-format and its unique identifier in the ERP database. Each client may have multiple sites where bins are actually located.
- ClientAddress is composed by the street name, the city name and the zip code in a human readable-format.
- ClientActivity/ClientActivityId corresponds to the client’s activity (Retail, Construction, Housing, etc.) in a human readable-format and its unique identifier in respect with the European regulation (NACE code) in the ERP database.
- Site
- SiteName/SiteNameId refers to the site’s name in a human readable-format and its unique identifier in the ERP database.
- SiteAddress is composed by the street name, the city name and the zip code in a human readable-format.
- SiteObservation/SitePreference provides any site information (key, phone number, etc.) and service preference such as the preferred time window (starting/ending time in the morning or in the afternoon, day or time-slot when collection is prohibited, etc.).
- Activity
- Each activity (bin emptying, bin exchange from an initial volume to a new one, etc.) is defined by a unique code in the ERP database, and that for each stage of the execution (from the command, to the intervention on site and finally when the service is done).
- WasteType provides the type of waste in a human readable-format that is planned to be collected. It enables filtering waste collection activities (e.g., domestic, glass or cardboard).
- WasteVolume refers to the bin volume (maximal capacity).
- WasteWeight provides the collected weight.
- WasteTimestamps corresponds to the service date.
4.2. WinFleet
- Time
- The timestamp of the entry.
- Status
- Information on the individual truck’s state:
- Is the motor running?
- Is the truck stopped or is it driving?
- How fast is the truck moving?
- Position
- GPS-information where this entry was taken:
- Latitude.
- Longitude.
- Street name in a human readable-format.
4.3. Sensors
4.3.1. Ultrasonic Sensing
4.3.2. Sensor Integration
4.3.3. Network Communication
4.3.4. Metrics
- Time
- The timestamp of the entry.
- Distance
- The distance (in cm) from the sensor to the nearest obstacle. This makes it possible to deduce:
- The filling level of the bin (depending on the dimension of bin).
- Alerts based on critical fill levels.
- Temperature
- The temperature of the bin, in degrees Celsius, for:
- Fire detection.
- Weather information retrieval.
- Battery
- The power level of the sensor, which are designed with a 5-year, interchangeable battery (according to the manufacturer’s information).
5. Data Analytics
5.1. ERP: DIVALTO
5.2. Winfleet
5.3. Sensors
6. Discussion
6.1. Data Aggregation–Correlation
- The real number of processed bins. It can be retrieved with ultrasonic sensors prior to the waste collection or at the end of the day when interventions are updated into the ERP database.
- The site configuration (indoor/outdoor, doors, access ramp, etc.). Unfortunately the distance and the time needed to travel from the truck to the bin location (and vice versa) are not known.
6.2. Optimization
- Too early: the truck arrives too early and the bins are not filled to an appropriate level. This is not desirable for the waste collection provider because the company invested its resources in a service which could be performed at a later point in time.
- Too late: the truck arrives too late and the bins are filled to overflowing. This is not desirable for for both the client and the waste collection provider. The waste collection provider now needs to invest more service time to collect all the waste and the client is unsatisfied with the provided service quality.
- Just in time: the truck arrives and the bins are filled to an appropriate level. This is the desired scenario.
- Optimization based on static information (see Section 5.1) from client contracts (e.g., allocation and clustering of clients by using their site’s location and their collection frequency).
- Expectation: This first optimization should significantly reduce the traveling distance to complete waste collection tours. However this step should have no influence on reaching the just in time scenario.
- Optimization based on added historical data to form client profiles and predict their waste generation (see Section 5.2).
- Expectation: With the use of historical data about the amount of waste collected at each client, it should be possible to identify weekly and/or seasonal patterns for waste generation. The accuracy of the resulting estimations needs to be examined, but overall this should lead to a significant improvement compared to the first step.
- Optimization based on added sensor measurement data (see Section 5.3).
- Expectation: Through the use of the near real-time measurements of the bins’ filling level, the accuracy of waste generation prediction can be improved. This should lead to further performance improvements compared to the second step.
7. Conclusions and Future Work
Author Contributions
Funding
Conflicts of Interest
Abbreviations
DR | Dynamic Routing |
DS | Dynamic Scheduling |
ERP | Enterprise Resource Planning |
GPS | Global Positioning System |
IOT | Internet Of Things |
MDPI | Multidisciplinary Digital Publishing Institute |
NACE | Socio-professional category |
PAM | Partitioning Around Medoids |
PCA | Principal Component Analysis |
PoL | Local waste company |
SWC | Smart Waste Collection |
SWAM | Smart WAste Collection SysteMs |
SWAM-DMS | SWAM Data Management System |
References
- How the Internet of Things Is Aiding the Garbage Crisis. Available online: https://newsroom.cisco.com/feature-content?articleId=1757267 (accessed on 6 February 2020).
- The Impact of Digital Transformation on the Waste Recycling Industry. Available online: https://store.frost.com/the-impact-of-digital-transformation-on-the-waste-recycling-industry.html (accessed on 6 February 2020).
- Sensoneo Smart Waste Management. Available online: https://sensoneo.com/ (accessed on 6 February 2020).
- EcubeLabs. Available online: https://www.ecubelabs.com/solar-powered-trash-compactor/ (accessed on 6 February 2020).
- Bigbelly Smart Solutions for Cities. Available online: https://bigbelly.com/ (accessed on 6 February 2020).
- SWAM Smart Waste Collection Systems. Available online: https://www.list.lu/en/research/project/swam/ (accessed on 6 February 2020).
- Faye, S.; Melakessou, F.; Mtalaa, W.; Gautier, P.; AlNaffakh, N.; Khadraoui, D. SWAM: A Novel Smart Waste Management Approach for Businesses using IoT. In Proceedings of the 6th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation (BuildSys 2019), New York, NY, USA, 13–14 November 2019. [Google Scholar]
- Faye, S.; Melakessou, F.; Gautier, P.; Khadraoui, D. Demo Abstract: No Time to Waste, with SWAM! In Proceedings of the 6th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation (BuildSys 2019), New York, NY, USA, 13–14 November 2019. [Google Scholar]
- Beliën, J.; De Boeck, L.; Van Ackere, J. Municipal solid waste collection and management problems: A literature review. Transp. Sci. 2012, 48, 78–102. [Google Scholar] [CrossRef]
- Laporte, G.; Osman, I.H. Routing problems: A bibliography. Ann. Oper. Res. 1995, 61, 227–262. [Google Scholar] [CrossRef]
- Srikantha, N.; Moinuddin, K.; Lokesh, K.S.; Narayana, A. Waste Management in IoT-Enabled Smart Cities: A Survey. Int. J. Eng. Comput. Sci. 2017, 6, 21507–21512. [Google Scholar] [CrossRef]
- Pardini, K.; Rodrigues, J.; Kozlov, S.; Kumar, N.; Furtado, V. IoT-based solid waste management solutions: A survey. J. Sens. Actuator Netw. 2019, 8, 5. [Google Scholar] [CrossRef] [Green Version]
- Catania, V.; Ventura, D. An approch for monitoring and smart planning of urban solid waste management using smart-M3 platform. In Proceedings of the 15th Conference of Open Innovations Association Fruct, St. Petersburg, Russia, 21–25 April 2014; pp. 24–31. [Google Scholar]
- Aazam, M.; St-Hilaire, M.; Lung, C.H.; Lambadaris, I. Cloud-based smart waste management for smart cities. In Proceedings of the 2016 IEEE 21st International Workshop on Computer Aided Modelling and Design of Communication Links and Networks (CAMAD), Toronto, ON, Canda, 23–25 October 2016; pp. 188–193. [Google Scholar]
- Ramasami, K.; Velumani, B. Location prediction for solid waste management—A Genetic algorithmic approach. In Proceedings of the 2016 IEEE International Conference on Computational Intelligence and Computing Research (ICCIC), Tamil Nadu, India, 15–17 December 2016; pp. 1–5. [Google Scholar]
- Saha, H.N.; Auddy, S.; Pal, S.; Kumar, S.; Pandey, S.; Singh, R.; Singh, A.K.; Banerjee, S.; Ghosh, D.; Saha, S. Waste management using Internet of Things (IoT). In Proceedings of the 2017 8th annual industrial automation and electromechanical engineering conference (IEMECON), Bangkok, Thailand, 16–18 August 2017; pp. 359–363. [Google Scholar]
- Thakker, S.; Narayanamoorthi, R. Smart and Wireless Waste Management-An innovative way to manage waste and also produce energy. In Proceedings of the IEEE Sponsored 2nd International Conference on Innovations in Information Embedded and Communication Systems (ICIIECS’15), Coimbatore, India, 19–20 March 2015; pp. 59–67. [Google Scholar]
- Folianto, F.; Low, Y.S.; Yeow, W.L. Smartbin: Smart waste management system. In Proceedings of the 2015 IEEE Tenth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP), Singapore, 7–9 April 2015; pp. 1–2. [Google Scholar]
- Malapur, B.; Pattanshetti, V.R. IoT based waste management: An application to smart city. In Proceedings of the 2017 International Conference on Energy, Communication, Data Analytics and Soft Computing (ICECDS), Chennai, India, 1–2 August 2017; pp. 2476–2486. [Google Scholar]
- Anagnostopoulos, T.; Zaslavsky, A.; Medvedev, A. Robust waste collection exploiting cost efficiency of IoT potentiality in Smart Cities. In Proceedings of the 2015 International Conference on Recent Advances in Internet of Things (RIoT), Singapore, 7–9 April 2015; pp. 1–6. [Google Scholar]
- Shyam, G.K.; Manvi, S.S.; Bharti, P. Smart waste management using Internet-of-Things (IoT). In Proceedings of the 2017 2nd International Conference on Computing and Communications Technologies (ICCCT), Chennai, India, 23–24 February 2017; pp. 199–203. [Google Scholar]
- Mes, M.; Schutten, M.; Rivera, A.P. Inventory routing for dynamic waste collection. Waste Manag. 2014, 34, 1564–1576. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Niska, H.; Serkkola, A. Data analytics approach to create waste generation profiles for waste management and collection. Waste Manag. 2018, 77, 477–485. [Google Scholar] [CrossRef] [PubMed]
- Divalto Solution. Available online: https://www.divalto.com/en/ (accessed on 6 February 2020).
- Winfleet Solution. Available online: https://www.skycom.com/ (accessed on 6 February 2020).
- BrighterBins Solution. Available online: https://www.brighterbins.com/ (accessed on 6 February 2020).
- Vejlgaard, B.; Lauridsen, M.; Nguyen, H.; Kovács, I.Z.; Mogensen, P.; Sorensen, M. Coverage and capacity analysis of sigfox, lora, gprs, and nb-iot. In Proceedings of the 2017 IEEE 85th vehicular technology conference (VTC Spring), Sydney, Australia, 4–7 June 2017; pp. 1–5. [Google Scholar]
- Hartigan, J.; Wong, M. Algorithm AS 136: A K-means clustering algorithm. J. R. Stat. Soc. Ser. C 1979, 28, 100–108. [Google Scholar] [CrossRef]
- Kaufman, L.; Rousseeuw, P.J. Finding Groups in Data: An Introduction to Cluster Analysis; John Wiley: Hoboken, NJ, USA, 1990. [Google Scholar]
- Al-Naffakh, N. A comprehensive evaluation of feature selection for gait recognition using smartwatches. Int. J. Inf. Secur. Res. 2016, 6. [Google Scholar] [CrossRef]
- Al-Naffakh, N.; Clarke, N.; Li, F.; Haskell-Dowland, P. Unobtrusive gait recognition using smartwatches. In Proceedings of the 2017 International Conference of the Biometrics Special Interest Group (BIOSIG), Darmstadt, Germany, 20–22 September 2017; pp. 1–5. [Google Scholar]
- Al-Naffakh, N.; Clarke, N.; Dowland, P.; Li, F. Activity recognition using wearable computing. In Proceedings of the 2016 11th International Conference for Internet Technology and Secured Transactions (ICITST), Barcelona, Spain, 5–7 December 2016; pp. 189–195. [Google Scholar]
- GraphHopper Solution. Available online: https://www.graphhopper.com/ (accessed on 6 February 2020).
- Geocoder Solution. Available online: https://geocoder.readthedocs.io/ (accessed on 6 February 2020).
Property | Public Waste Collection | Commercial Waste Collection |
---|---|---|
Client scope | Private households | Companies, restaurants, schools |
Client waste generation | Homogeneous | Heterogeneous (1000–10,000L per week) |
Collection scope | Communities | Country-wide/regions |
Distance between clients | Short (neighbors) | Significant, but can be diverse |
Bin placement | On the pavement | Somewhere on the client site |
Study | Bin Location | Waste Type | Sensors | GPS | Implementation | DS | DR |
---|---|---|---|---|---|---|---|
[13] | Outdoor | Glass; Plastic; Paper; General Waste | Weight & Capacity | No | Simulation | No | No |
[14] | Outdoor | Organic, Plastic/ Paper/Bottle, Metal | Capacity | No | Simulation | Yes | Yes |
[15] | Outdoor | General | - | Yes | Simulation | No | No |
[16] | Outdoor | General | Capacity | No | Simulation | Yes | Yes |
[17] | Outdoor | Plastic | Weight & Capacity | No | Simulation | No | No |
[18] | Outdoor | General | Capacity | Yes | Real | No | No |
[19] | Outdoor | General | Capacity | No | Simulation | Yes | Yes |
[20] | Outdoor | General | Capacity | Yes | Simulation | Yes | Yes |
[21] | Outdoor | General | Capacity | Yes | Simulation | No | Yes |
[22] | Indoor | General | Capacity | No | Real | Yes | Yes |
[23] | Outdoor | General | Weight | Yes | Real | No | No |
Activity | Cluster 1 | Cluster 2 | Cluster 3 | Cluster 4 |
---|---|---|---|---|
Retail Business | 0.302 | 0.581 | 0.023 | 0.093 |
Construction | 0.531 | 0.406 | 0.031 | 0.031 |
Housing Activity | 0.400 | 0.300 | 0.200 | 0.100 |
Restaurant | 0.185 | 0.481 | 0.074 | 0.259 |
Activity | Cluster 1 | Cluster 2 | Cluster 3 | Cluster 4 |
---|---|---|---|---|
Retail Business | 0.279 | 0.511 | 0.093 | 0.116 |
Construction | 0.531 | 0.219 | 0.219 | 0.031 |
Housing Activity | 0.400 | 0.000 | 0.300 | 0.300 |
Restaurant | 0.185 | 0.222 | 0.296 | 0.296 |
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Melakessou, F.; Kugener, P.; Alnaffakh, N.; Faye, S.; Khadraoui, D. Heterogeneous Sensing Data Analysis for Commercial Waste Collection. Sensors 2020, 20, 978. https://doi.org/10.3390/s20040978
Melakessou F, Kugener P, Alnaffakh N, Faye S, Khadraoui D. Heterogeneous Sensing Data Analysis for Commercial Waste Collection. Sensors. 2020; 20(4):978. https://doi.org/10.3390/s20040978
Chicago/Turabian StyleMelakessou, Foued, Paul Kugener, Neamah Alnaffakh, Sébastien Faye, and Djamel Khadraoui. 2020. "Heterogeneous Sensing Data Analysis for Commercial Waste Collection" Sensors 20, no. 4: 978. https://doi.org/10.3390/s20040978
APA StyleMelakessou, F., Kugener, P., Alnaffakh, N., Faye, S., & Khadraoui, D. (2020). Heterogeneous Sensing Data Analysis for Commercial Waste Collection. Sensors, 20(4), 978. https://doi.org/10.3390/s20040978