STAY CONNECTED WITH US
    

admin@infinitbis.co.za | +27 83 647 6773 | +27 86 206 0502

WHERE SAVING LIVES IS A DAILY HABIT

Slider 2 SILO: ENTERPRISE SILO: ENTERPRISE RISK MANAGEMENT SYSTEM
Slider 1 ARTIFICIAL INTERLLIGENCE ARTIFICIAL INTERLLIGENCE ARTIFICIAL INTERLLIGENCE
Slider 3 PREDICTIVE ANALYSIS PREDICTIVE ANALYSIS
Slider 4 RAPID APPLICATION DEVELOPMENT RAPID APPLICATION DEVELOPMENT
Slider 5 MOBILITY MOBILITY
Slider 6 IMPOSSIBLE IS ONLY A OPTION

Silo Analytics

All processes in this module occur in real-time, with results rendered the moment a change occurs in the client environment. Results are typically disseminated through the Silo-t module, if the derived intelligence indicates an immediate or serious impact on core business, or as a series of dynamic dashboards or automatically generated, intuitive reports submitted on a client-determined schedule.

The analytical layer supports a wide variety of algorithms ranging from basic operational analysis capable of identifying recurring threats increasing trends in risk profiling or even workplaces or work teams with an exceptional tendency for risk events; to advanced prescriptive models such as cause-event analysis (identifying leading indicators for future events based on current conditions), behavioural impact (behaviour in the workspace leading to unacceptable trends or occurrences) and predicted impact of current events (predict the 6 month impact of an accident). Silo-a transforms Silo from a sophisticated management system into a cutting edge risk intelligence solution capable of completely transforming your enterprise risk landscape.

 Statistical and trend analysis
 Algorithm  Execution context Description  Example 
Trend analysis based on
mathematical behaviour

 Real-time

 

Application of standard statistical models to a series of recorded values to attempt identification of the governing mathematical rule controlling occurrence of said values

CO2 levels in a specific area measured over time, and trended to predict behaviour

Trend analysis based on pattern identification

 Real-time

Identification of logical patterns governing the occurrence of a series of recorded values. These patterns are non-mathematical, and may be based on influence by eternal data sets

Number of first aid injuries as trended against wage or salary pay-out data

Ratio calculation and monitoring

 

 Real-time

 

 

Comparison of any two mathematical values derived from database queries, to each other – the one as a fraction of the other. The relationship may also be subject to an upper and/or lower value barrier

Ratio of male to female employees, measured inside legislative boundaries

Bracketing

 

 

 

Scheduled,
ad-hoc

 

 

Identification of benchmark bands or value brackets based on summation and historical analysis

 

 

 Financial impact of injuries sustained to a victims' head typically falls between R10 000 and R20 000

Benchmarking and measurement

 

 

Real-time

 

 

 

Comparison of recorded values to a set of pre-configured business rules or sourced references, with the intent to identify the measure of deviation from this reference value

Compare injury frequency rates for mining operations against the rates of other members of the industry


Descriptive analytics
 Algorithm  Execution context Description  Example 

Relationships

 

Real-time, scheduled and ad-hoc

 

Identify the manner in which different data sets have an impact or association with another set of data blocks

Identify the manner in which specific event causes relate to the injuries sustained during these events

Timelines

 

 

Real-time, scheduled and ad-hoc

 

Analyse the temporal aspect of data blocks in an attempt to identify how time describes or impacts on said data block

Describe the days of the month on which production has a tendency to drop significantly

Habits

 

 

Scheduled

 

 

Identify the occurrence of specific data blocks in a sequence or order that is repeated, or repeated closely enough so as to constitute a possible

Identify the manner in which risk officers complete their safety audits

           

Populations

 

 

 

 

 

Real-time, scheduled

 

 

 

 

 

Identify data populations based on classification and implied grouping derived form the data blocks in which these populations reside. Analyse these populations‟ behaviour to identify both typical and anomalous behaviour based on population majorities

Profile the throughput for a set of conveyor structures based on location and ore mined, and identify any that are performing 20% less than the typical members.

 

Conjunction

 

 

Real-time, scheduled

 

 

Analyse the manner in which the operational timelines of identified data entities impact on one another, and how these impacts affect or describe the perceived behaviour of these entities

Identify which guards are on duty whenever a certain amount of ore is stolen from a smelter.

 

Compliance

 

 

 

Real-time

 

 

 

Analyse the manner in which a specific operational sequence or process is adhered to (compliance) or deviated from (deviation) as measured against a preconfigured procedure

Identify rock engineers who do not follow the prescribed process of testing as outlined in company operating procedures

Data quality

 

 

 

Real-time

 

 

 

Monitor the completeness of data blocks against configured quality rules in an effort to identify data sets that, through being incomplete or wrong, have a negative impact on Silo's operation

Identify investigations that have not been completed in an appropriate fashion

 

Impact

 

 

Real-time, scheduled

 

 

Identify typical business impacts related to especially unwanted events, and build a neural network codifying this relationship

Identify the typical legal impact of a fall of ground event

 

Business rules

 

 

 

Real-time

 

 

 

 

Apply recorded business rules to data blocks, as specified by the definition or implied references of said business rule

 

 

Identify potential reportable injuries based on timesheet behaviour of the injured, and a set of business rules outlining the definition of a reportable injury

Modelling

 

 

 

Real-time, scheduled

 

 

 

Apply and compare defined data models against data blocks recorded in Silo

 

 

Identify the bonus bracket into which a specific mine manager fits based on productivity and other modelled performance indicators

 

Predictive and prescriptive analytics
 Algorithm  Execution context Description  Example 

Event postulation

 

 

Real-time

 

 

 

Consume different aspects of the descriptive analytical ability to postulate the occurrence of certain future events

 

Postulate the occurrence of a future fall of ground event based on current rock engineering, safety and operational assessments and measurement

Guidance

 

 

Real-time, scheduled

 

 

 

Describe the change in behaviour of a set of data blocks or data entities subsequent to an undefined trigger event or change

Advise which series of actions have the most positive influence on the resolution of a section 54 instruction

 

 Modelling

 

 

Real-time, scheduled and ad-hoc

 

Describe data model most suited to a specified operational result through extrapolation of historical data blocks

Advice which combinations of personnel in which mining teams yield the highest production indicators

 Mapping

 

Real-time, scheduled and ad-hoc

 

Describe the extrapolated environment for a specific discipline based on historical reference

Extrapolate and describe the risk environment for a specific shaft across the following 6 months