Jumpstarting an Integrated Township Operations Center (Smart City) Using People as Sensors
0 likes641 views
The document discusses the concept of integrated townships that encompass various domains such as traffic, water, energy, public safety, and health. It proposes leveraging citizen input as a cost-effective and timely alternative to traditional sensor data for monitoring these domains, facilitating quicker decision-making and resource optimization. The approach aims to enhance data integration while physical sensors are developed, ultimately achieving a smarter urban environment.
![© 2015 IBM Corporation
Helping build a smarter planet
Format for Entering Information
§ Location: [Pre-defined list of city regions auto-completed based on first-few letters]
§ Domain: [Traffic, Water, Energy, Public Safety, Health]
§ Status: [Red, Yellow, Green, Unknown] + <Optional data>
– Optional data can be speed in traffic, kw-hour in energy, etc.
– Depending on device capability, it can be audio, image or video as well.
4](https://image.slidesharecdn.com/projectidea-discussion-bits-mar2015-150502104824-conversion-gate02/85/Jumpstarting-an-Integrated-Township-Operations-Center-Smart-City-Using-People-as-Sensors-5-320.jpg)
Jumpstarting an Integrated Township Operations Center (Smart City) Using People as Sensors
- 1. © 2015 IBM Corporation Integrated Township Operations Center Jump-started by People as Sensors Biplav Srivastava, Ullas Nambiar, Vikas Agarwal, Sumit Mittal Dec 2010 Instrumented Interconnected Intelligent Smarter
- 2. © 2015 IBM Corporation Helping build a smarter planet Motivation and Problem § An integrated township has elements of multiple domains like traffic, water, energy, public safety and health. § Benefit Possible: • Getting a composite view helps administrators make timely decisions • Helps optimize resources across domains (Smarter cities value) § Problem: • Getting data from instrumentation is costly, slow, noisy, insufficient and error-prone • Once data is obtained, getting interconnection right is complicated • Getting integration across domains is hard, given the challenges with physical sensors • Time to value is too long and subject to too many variables, including physical limitations 1
- 3. © 2015 IBM Corporation Helping build a smarter planet Solution Idea and Advantages § Idea – Ask people to give information about sensing information in different domains in a uniform, structured manner – Exact affected domains can be inferred – Status of individual and across domains is updated – Jumpstarts integration while physical sensors are integrated in parallel § Advantages – Expedites time to value – Improves acceptance of solution – Integration issues are removed while physical sensors are tested/ integrated separately 2
- 4. © 2015 IBM Corporation Helping build a smarter planet Mockup 3 City: <My Town> Date: <6 Sep 2010> Time: <Local Time> Name: <Manish Nambiar> Role: <Town Mayor> Town Monitor Safety Traffic Energy Water Health Overall Situation: • Water leakage in some region • People are agitated • Electricity has to be cut off to help repairs • Sanitation issues • Traffic unaffected Town Context: • Class 2 city in India • State utilities provide basic amenities • Limited budget, need to show early results
- 5. © 2015 IBM Corporation Helping build a smarter planet Format for Entering Information § Location: [Pre-defined list of city regions auto-completed based on first-few letters] § Domain: [Traffic, Water, Energy, Public Safety, Health] § Status: [Red, Yellow, Green, Unknown] + <Optional data> – Optional data can be speed in traffic, kw-hour in energy, etc. – Depending on device capability, it can be audio, image or video as well. 4
- 6. © 2015 IBM Corporation Helping build a smarter planet Uniform Manner of Asking Inputs from Crowd and Inference Unknown (Gray) Green Yellow Red Unknown (Gray) Unknown (Gray) Unknown (Gray) Unknown (Gray) Unknown (Gray) Green Unknown (Gray) Green Yellow Red Yellow Unknown (Gray) Yellow Yellow Red Red Unknown (Gray) Red Red Red 5 Levels for status for a domain in highest to lowest order: • 4-level example: Unknown, Green, Yellow, Red • Semantic based on domain • Semantics for traffic = unknown speed, free flow, slow traffic and congested traffic • Semantics for water = unknown pressure, high pressure, medium pressure and no/ low pressure • Semantics for energy= unknown supply, high availability, supply limited and power outage • Semantics for safety= unknown, no incidents, minor incidents and major incident(s) reported • Semantics for health= unknown status, hospital beds available, beds limited and no bed availability Inferring composite status across domains: • Conservative aggregation: If level-1 and level-2 are status levels for two domains, composite level = min(level-1, level-2) • Majority aggregation: aggregate level is the level of the maximum number of domains Illustration of conservative aggregation
- 7. © 2015 IBM Corporation Helping build a smarter planet Some Scenario Details Actors Information uploaded Information sent Safety Citizens, Police personnel Text (incident), Video, Audio Text (advisory) Traffic Citizens, Municipal workers, Police personnel Text (speed, condition), Video, Audio Text (advisory) Energy Citizens, Energy utility personnel Text (outage) Text (advisory) Water Citizens, Water utility personnel Text (outage) Text (advisory) Health Citizens, Hospital contact Text (condition) Text (advisory) Overall Administrative heads, emergency management personnel Text (instructions), Video, Audio Text (advisory) 6