Water Interruption Forecasting

 Capstone Project Proposal #3

I.     Proposed Topic

Water Interruption Forecasting

II.     Introduction

Fresh water is one of the scarce resources today.  Among all the water on Earth, only 1% is fresh water usable for human consumption. Currently many countries are facing water problem due to population and economy growth, especially in large cities. (Dai, Li, Sahu, Naphade, & Chen, 2011)

Water utilities are mandated to deliver safe and potable drinking water to its franchise area.  Inefficiency in the quality and quantity of drinking water is detrimental to the operations of the water utility and the overall health of its concessionaires.  Measures should always be taken and practiced to deliver potable drinking water 24 hours a day for it is said that a person can go on for days without food but not without water.

Unfortunately, no matter how prepared a water service provider in terms of delivering potable water to its area, water loss or inadequacy thereof is still bound to happen causing incessant rantings from affected consumers.  In addition, water loss would greatly affect the revenues of the water utility as this would be considered a financial waste.

Hence to avoid inconvenience to the public in general, a system will be developed that will forecast water interruption in order to apprise the water utility of the matter and prevent or mitigate the effects thereof.  This will be done by data mining all cases of water interruption the water utility had experienced, whether it be from natural or man-made causes such as cases of preventive maintenance, water loss due to leakage, low pressure experienced in the water distribution system, pump motor repairs and others.

III.     Purpose of the Study

The purpose of this project is to develop a system that will forecast water interruption.  The objectives are as follows:

  1. To prevent or mitigate the effects of water interruption by finding ways to compensate for the water loss or low water pressure;
  2. To inform the consumers of the imminent water loss;
  3. To prepare maintenance works after the water interruption.

IV.     Specific Research Issues/Questions

This project arises from the following issues:

  1. Does the water utility inform, in advance, their concessionaires regarding the possible water loss or low water pressure?
  2. If water interruption is experienced, for how long is the duration until the normal operations of the water utility is restored?
  3. Do mechanisms are in place to detect impending water interruption?

These issues will be discussed below.

  1. Does the water utility inform, in advance, their concessionaires regarding the possible water loss or low water pressure?

Proper dissemination of information is very important especially if it concerns a primary resource’s loss or lack thereof.  Water is as vital to life as in the air we breathe – it is used for drinking, cooking, cleaning, washing dishes and clothes, watering of plants and others – thus it is considered a “mortal sin” for any water utility to cut-off the water supply without reason and early notification.

In Digos Water District, concessionaires are informed if preventive maintenance will be conducted.  Preventive maintenance in the DWD setting includes the following:

  • Cleaning and disinfecting of reservoirs;
  • Pump and/or motor pull-out for efficiency testing and other maintenance works;
  • Flushing activities; and
  • Major pipeline installation or rehabilitation.

On the average, information is made two (2) days prior the actual activity in order to give ample time for concessionaires to prepare for the imminent water loss or low water pressure.  Information drive were done thru radio broadcasts, crawler ads in the local cable TV provider, in the Digos Water District webpage and in DWD Facebook account.

In other water districts such as Davao City Water District, the same information drive was likewise conducted in order to apprise the consumers of the matter.

Nevertheless, on extreme emergency cases such as pipe burst which is difficult to predict, information on water interruption will be made the soonest time possible to affected areas although most of the time, it is the concessionaires who had the knowledge of the event and would just notify the water district of the problem.

  1. If water interruption is experienced, for how long is the duration until the normal operations of the water utility is restored?

In the Digos Water District situation, available data on the average duration of water interruption is set as follows:

Activity resulting to water loss

Estimated size of affected area

Average duration of water loss

Flushing Around 50-100 households 4 hours per area and conducted between 12:00AM – 4:00AM
Major pipeline installation such as installation on a bridge Minimal, since the district will ensure that affected areas will be supplied with water 3 days

As noted, there is no data on other preventive maintenance activities such as pump pull-outs and cleaning of reservoirs because most of the time, whenever these activities will be conducted, DWD made sure that there is enough supply to nearby areas although it can be said that low water pressure is experienced throughout the activity’s duration.

In cases of pipe burst which is currently unpredictable, depending on the gravity of the burst and the size of the affected areas, the maintenance crew of the of DWD will be directed to leave their usual assigned task (if it is not that urgent) to help in the repair of the pipe burst until it is restored.  Unfortunately, there is no concrete data that will show the duration of the water loss as the event varies from time to time.

  1. Do mechanisms are in place to detect impending water interruption?

Presently, there is no mechanism that will detect impending water interruption in Digos Water District.  Scheduling method is the sole basis of water loss or low water pressure information – absent any schedule, normal operations in the water district is presumed.

V.     Proposed Methodology

  1. The existing system of Digos Water District will be mined for any data in relation to complaints for water loss or low water pressure. In addition, notices for water interruption will be collated in order to obtain the necessary data.  The following data will be mined:
    • Reason for water interruption;
    • Location of water interruption;
    • Estimated size of households affected;
    • Actual complaints received in relation to the water interruption (in order to apprise the water district of the extent of information campaign previously made);
    • Duration of the water interruption;
    • Post activities following water interruption.
  2. A database will be made in relation thereof.
  3. A system will be developed that will show the above-stated data and a forecast of the likelihood of its occurrence in the future.

VI.     List of Readings

Literature on Behavior and Interruption

The works of Adamczyk, Iqbal, & Bailey, (2005) is quite far in concept as juxtaposed to this project but nevertheless it shows the behavior of a person when confronted with interruption.  It said that, “when applications interrupt a user at an inopportune moment during task execution, the user performs tasks slower, commits more errors, makes worse decisions, and experiences more frustration, annoyance, and anxiety than if it had interrupted at a more opportune moment.  Unfortunately, systems that attempt to manage human attention largely lack the facilities to make reasoned or even informed decisions about when to provide new information to users.”  In concluding their paper, the “results showed that predicted best points for interruption consistently produced less annoyance, frustration, and time pressure, required less mental effort, and were deemed by the user, more respectful of the their primary task.”

Furthermore, this theory was used as a complimentary approach in the paper of Gluck, Bunt, & McGrenere, (2007) which “examined the effect of timing of interruption onset to determine if negative effects can be reduced by presenting an interruption at an ideal moment and postponing the interruption if the moment is inopportune.”

Following the theory enunciated above, generally, interruptions are annoying and unwelcome but if measures were taken to make early information for imminent interruptions, this would be the opportune time so that people would be able to prepare beforehand and decrease or reduce annoyance level among the concessionaires.

Works on Forecasting

There had been an abundance of forecasting applications in the computing world.  To cite a few, there is the forecasting tool on a mobile device that focuses on casual users (Mercier, Dupin, Ulmer, & Demund, 2011); predicting water consumption to motivate conservation in residential households that consume a large portion of cities’ water supply and providing a heat-map for multi-resolution prediction (Dai et al., 2011); and weather forecasting using a deep hybrid model (Grover, Kapoor, & Horvitz, 2015).

Unfortunately, the proponent had noticed that forecasting on water interruption or any subject or topic in relation thereto is devoid in literature on the internet.

VII.     Expected Significance of the Study

The proponent admits of the limitation of the extent of the proposed project – limited to the existing data a water utility has over the years.  This is more or less equated to forecasting when the earthquake dubbed as the “Big One” would occur.  However, the proponent is positive that this would be more or less concise in the future and would be forecasting in the same vein as predicting the weather.

Relative to the water utilities, this project would be very helpful as it minimizes the complaints received and on the business side, it would give the water utilities other remedy in order to even out possible losses in revenue.  Furthermore, granting that the water utility had informed the concessionaires in advance, the concessionaires would be prepared when the water interruption will be experienced.

This project may become a source for future study about forecasting water interruption since the proponent had found out that there is a very limited literature on the matter.

VIII.     References

Adamczyk, P. D., Iqbal, S. T., & Bailey, B. P. (2005). A Method, System, and Tools for Intelligent Interruption Management. In Proceedings of the 4th International Workshop on Task Models and Diagrams (pp. 123–126). New York, NY, USA: ACM. http://doi.org/10.1145/1122935.1122959

Dai, J., Li, M., Sahu, S., Naphade, M., & Chen, F. (2011). Multi-granular Demand Forecasting in SmarterWater. In Proceedings of the 13th International Conference on Ubiquitous Computing (pp. 595–596). New York, NY, USA: ACM. http://doi.org/10.1145/2030112.2030230

Gluck, J., Bunt, A., & McGrenere, J. (2007). Matching Attentional Draw with Utility in Interruption. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 41–50). New York, NY, USA: ACM. http://doi.org/10.1145/1240624.1240631

Grover, A., Kapoor, A., & Horvitz, E. (2015). A Deep Hybrid Model for Weather Forecasting. In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 379–386). New York, NY, USA: ACM. http://doi.org/10.1145/2783258.2783275

Mercier, O., Dupin, S., Ulmer, C., & Demund, J. (2011). Forecasting Tool on a Mobile Device. In Proceedings of the 23rd Conference on L’Interaction Homme-Machine (pp. 7:1–7:8). New York, NY, USA: ACM. http://doi.org/10.1145/2044354.2044363

IX.     GANTT Chart

3 Water Interruption

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s