Artificial intelligence assisted ditch improvement plan
Artificial intelligence assisted ditch improvement plan
Track 02: Artificial intelligence, analytics, and value from data
Tags: artificial intelligence, visualization, APIs, GIS, water system monitoring, water quality.
Sponsor: Metsä Group
Challenge
The objective is to create a demonstration of a model that would allow an artificial intelligence assisted software to prepare a ditch improvement plan with the help of a digital terrain model (and other data).
Background
Ditch improvement enhances drainage, secures the growth of trees and facilitates the regeneration of forests. At the same time, it reduces the risk of phosphorus leaching resulting from a rise in the groundwater level. The need for ditch improvement is determined on the basis of the groundwater level, which should be at a depth of 30–50 cm during the growing season of trees. One of the indicators of a high groundwater table and poor drainage is an abundance of peat moss.
The sound planning of ditch improvements in forestry is extremely important for water protection. The main goal of ditch improvement with regard to water protection is to minimise scouring in the ditches to be cleaned. Some 70–90 per cent of the solid matter carried along with water should be stopped before the waters of the ditch empty into waterways. The leaching of solid matter and nutrients is minimised by leaving the ditches likely to be scoured uncleaned or by slowing down water flow in them in such a way that erosion does not take place. Water protection structures are dimensioned and placed in the ditch so that the movement of solids is prevented or so that the solid matter washed out is stopped effectively within the ditch.
Read the ditch improvement guide and develop a model that would enable the preparation of a ditch improvement plan – including the water protection measures – with the help of the artificial intelligence and a digital terrain model. The ditch improvement plan produced by the artificial intelligence would function as a basis for a terrain inspection carried out by a professional, who would serve to specify the plan.
Approaches and considerations
Consider what information is needed to design a ditch improvement plan. Analyze how a human does the task. Where is that information coming from? What is the basis for the planning decisions? What variables influence the results.
How to use software to perform these steps. You can use an automated algorithm, chain micro-services, or perhaps use machine learning to train the software. Explore different opportunities and compare them.
Build a demo and provide means to estimate the results and their reliability and validity.
Data
Collected data package
https://drive.google.com/drive/folders/1-oa5A6Mj87zPMEFMviSMr9DVzEGWPebX?usp=sharing
Technology
You can consider and try out the below examples and technologies. However, you’re not limited to use them at all, or only them. You should explore different approaches and find technology that supports your idea.
IBM DSX / Machine Learning
http://heidloff.net/article/watson-machine-learning-sample
https://datascience.ibm.com/community
Watson APIs
https://www.ibm.com/watson/developercloud/doc/index.html
https://www.ibm.com/watson/developercloud/starter-kits.html
Watson Visual Recognition
https://github.com/IBM-Bluemix/Visual-Recognition-Tile-Localization
ArcGIS Online of Xamk
Hackathon groups are granted an account for ArcGIS Online GIS platform. Your account has professional map authoring, publishing and analysis capabilities and rights.
A username and a password for the first login are provided per group by the staff.
Do NOT try to sign in here: http://www.arcgis.com/
The SIGN IN address is https://xamk.maps.arcgis.com/
Esri ArcGIS Desktop
You can use a trial desktop installation.
http://www.esri.com/arcgis/about-arcgis
Esri developer docs