Energy and Climate Change CommitteeSupplementary written evidence submitted by the Energy Services and Technology Association

FINAL DETAILED EVIDENCE FROM ESTA FOR DECC SELECT COMMITTEE, 24 JUNE 2013

	Issue	Resolve by
1	Divergence (see fig1). SMIP requirements are emerging faster than implementation is progressing implying we will never have a solution	Mandate use of generic protocols in DCC that are not application specific and do not need to be fixed against a time-line
2	Technical robustness—Program has relied on expertise from trade associations and IT (not meter) consultants without accountability for delivery. Little confidence on technical capability	Require vendors to prove capability and interoperability before DCC award
3	Exclusivity of access to Big 6 and their supply chain. Customer access only through what is granted in this chain.	Mandate proper accessible standards for meter communications
4	Onerous security requirements, impeding 1 and 2 and upholding 3	Remove supplier remote disconnect, and restrict this to DNO only
5	Displacement of many large existing systems and processes (eg ECOS, SCOGES, DTN, Elexon, ASPCop) without full consideration of implications	Review SMETS against existing systems, and assess why they are being replaced
6	Maximum coverage 70% of sites. No solution for remainder.	Determine a complete solution that can be mandated in SMETS before DCC award
7	Little control of existing Intellectual Property Rights of developed material and access to it outside contracted consultants—potential DCC procurement flaw	Require DCC bidders to outline what and how much they are relying on the knowledge of existing consultants/contractors moving over from DECC
8	Extremely low consumer engagement with a Program of this size—critical dependency for success	DECC (not supplier) funded awareness campaign that outlines burdens as well as benefits and how a good working metering market will ensure customer gets a good deal
9	Build-up of stranded assets of non-interoperable meters permitted to allow roll-out to start	Do not permit further non-interoperable meters to be installed, delay award of DCC until interoperable meters are available, or allow any meter that meets the minimum SMETS1 interoperable standard to count towards supplier obligations
10	Concern that SMIP will not deliver savings envisaged	Engage more environment and sustainability groups in developing further SMETS specifications

Fig 1

1. Divergence

See Fig 1. The flaw in current implementation design is the need to fix the specification at a point in order for detailed development to take place. For the foreseeable future, requirements for the SMIP are emerging faster than the GB Companion Specification (GBCS) can keep up. This implies that we will never have an interoperable solution using current methods. The GBCS must be complete to the fixed point in absolute detail before implementation can be made. This involves detailed analysis, discussion and agreement between competing parties, requiring the sharing of intricate design information of their individual products. ESTA made notes on issues from two recent SDAG meetings (firmware upload and prepayment) which were circulated to the members present and any comments taken in. The notes highlighted 20 red issues on which the GBCS is dependent. The original target completion date for the GBCS was April 2013. This was then moved to December 2013 and currently stands at April 2014. How confident are we the the time-line will not slip again?

DECC must take a firmer approach with vendors who want to have “control” of the application layer of the design. Both Zigbee and DLMS associations demand this “control”, unnecessarily because both support generic protocols (Zigbee PRO and DLMS/FLAG). The GBCS can be (and generally is) implemented through these generic protocols. This approach must be mandated, rather than specifying application data in detail. The latter can be achieved through bilateral dialogue between vendor and customer (energy supplier); the benefit of this approach is that the lack of completeness in the GBCS does not prevent the roll-out of DCC and Comms Hubs (CH). The CH can transmit generic messages between HAN and WAN without them needing to be defined completely. This definition can be achieved in the Head End System (HES), In Home Display (IHD) and Consumer Access Device (CAD) through the use of vendor specific “drivers”, similarly to the way a new “gadget” is connected to a PC. This will be a much more robust method than using complete firmware upgrades.

2. Technical Robustness

ESTA attended the first two Industry Day seminars run by DECC, where DCC (CSP and DSP) bidders were invited to present their proposed solutions. Members for current large scale data retrieval companies (Technolog and Siemens) attended and asked challenging questions about the technical implementation. Whilst bidders were strong on general communications and database systems there was an obvious lack of expertise in detailed meter protocol support. This aspect is essential in the SMIP because both the CH and HES depend on it. An added complication is because CH and HES are supported separately (by CSP and DSP) there will need to be arbitration between each technical solution, falling to the overall DCC bidders. Neither DCC bidder was able to demonstrate the technical expertise to perform this function. Whilst ESTA accept that technical expertise can be brought in at a later date, there are technical impositions here which will shortly become irreversibly fixed, outside the control of any bidder. ESTA has offered expertise to assist in reducing this risk, but were not invited to the third Industry Day in May. We have not yet been advised by DECC why we were not invited.

3. Exclusivity

The HAN and WAN are, and are likely to remain under the exclusive control of the Big 6 Energy Suppliers. ESTA are the only representing body on SDAG, HAG and CWG that represent the demand side. It addition ESTA have restricted access to many of these groups, in particular GBCS meetings. Note recent lobbying has resulted in some of the restrictions being lifted, but ESTA are still only an “observing” member despite providing valuable input to the design process. Whilst we appreciate the difficulty in getting coherent consumer representation, it is essential that the demand side position is taken in because this is what will be needed to engage the consumer. If exclusivity is upheld the consumer will simply not allow the SM to be installed (an option he now has). Furthermore if the consumer is forced to pay a premium if he does not have a smart meter, then surely he must be given a choice of what type of smart meter he can have (see 5 Displacement).

4. Onerous Security

The functions specified in SMETS have initiated concerns over security and the need for the implementation to be one of critical national infrastructure. However this is only the case if the supplier continues to retain the right to remotely disconnect using the current implementation model. The need for such security has impacted the Program significantly, and we are still uncertain about the effect on the details of the design in many areas. ESTA urge DECC to review whether this privilege is necessary for suppliers. They already have the facility to switch to prepayment mode. The only need for remote disconnect is to de-stress the network, and this is therefore a function only of the DNO. In addition it emerged recently that the meter is not a fiscal instrument, and double accounting will be performed (as at present). This negates need for such security for protection of financial information since they are for consumer indication only.

In current metering systems, where tariffs and configurations are programmed into meters, a PIN Sentry type of algorithm with “write-only” key is used, similar to that used on Internet banking. ESTA would ask DECC to review this method against the table below, using a number of use case scenarios, to take a pragmatic view of cost against risk, perhaps phasing in Method 3 gradually. ESTA would not recommend Methods 4 or 5.

DECC must take a firmer approach with suppliers on whether they should have the privilege of remote disconnect.

Method	Access to Remote Switch	Potential Breach and Remedy
1	SM includes remote switch but is not enabled.	None
2	Remote switch is enabled, but access code and security algorithm for each version is not passed outside meter vendor	Infiltration of access code and security to individual meter type from meter vendor. Security breach entirely with vendor. Identify breach and recall/replace meters affected
3	Individual meter vendors provide security access to activate remote switch for each meter version to each distribution business where such meters are installed	If infiltration is restricted to a single distribution area, then security breach is with that distribution company. If the infiltration is across multiple DNOs then breach is as 2
4	Same as 3, except information is provided to supplier rather than distributor	Review history of suppliers to meters with breach. If a single common supplier is identified then breach is with supplier. If a single common meter type is identified then breach is with vendor.
5	Single universal highly secure method. Access to DCC, DNO and suppliers	Difficulty identifying who is in breach. Complete replacement required

There is an additional complication mentioned at the recent prepayment meeting where the remote disconnect could be used by suppliers to “simulate” prepayment functions in SMETS1. Apart from the major implications this would have on the Program it would be unreasonable for the consumer, and he would be unlikely to accept it. Being switched off, without the possibility to switch to another supplier (in the case perhaps of a dispute) is a predicament DECC would not wish to impose on the consumer.

5. Displacement

ESTA are concerned that the SMIP will displace many metering systems already installed for the consumer to help reduce demand. Customers have paid for these metering systems, and ESTA members revenue depend on their continued operation. Whilst DECC assure ESTA that opt-out will be available to I&C consumers, there is significant risk of creep of the DCC, especially when its costs can be hidden inside an energy suppliers bill. It is likely too that the meter provided under the SMIP will have less specialised functionality than has been installed by the ESTA member.

There is also a general concern that existing systems are being overlooked. Some of these systems have been in place for 15 years, and have taken time to settle in. Replacing them with “green-field” designs exposes us to risk of repeating processes and making the same mistakes a second time. Such exposure is difficult to quantify but is likely to be significant. Such systems include The Data Transfer Network (DTN), ECOS for electricity meter details, SCOGES the equivalent in gas, the MRA (Master Registration Authority), Elexon for electricity balancing and settlement, the Quantum system for prepayment, CHIRPS for interoperability, and the ASPCop (AMR Service Provider Code of Practise). The latter is extremely important because it allows the consumer to nominate a smart meter provider, the readings from which the supplier is required to accept. This empowering of the consumer is essential for engagement and must be upheld.

6. Lack of Coverage

The SMIP currently mandates Zigbee at 2.4 GHz as the frequency for the HAN. However, tests that ESTA and other industrial stakeholders participated in showed that this will only properly address 70% of properties n the UK. Whilst there is provision in the SMETS2a specification for 868MHz where coverage will be vastly improved, we do not have certainty that the full GBCS can be implemented at 868 because of lower bandwidth. Going ahead at 2.4GHz which DECC propose will preclude consumers from these benefits if they cannot be implemented.

The use of wired connections (power line carrier, PLC) is precluded in SMETS2 because there is no mandated provision for the connections necessary. Vendors are merely “strongly advised” to include them.

Such decisions have made it difficult to generate sufficient interest from RF and PLC vendors to commit to developing product. Only recently has the Zigbee Market Requirement Document (MRD) for 868 been raised, and we are due a PLC industry day in September, where vendors will be showing their wares. However this does not give us assurance of completeness since the DCC contract is scheduled to be awarded before this.

The use of Zigbee at 2.4GHZ in SMETS2a was successfully notified to the EU on the 24th January. However the SDAG had already raised many issues on this over the Christmas period, and the SDAG meeting on 24th January was ear-marked to debate and resolve the outstanding issues before notification. However the SDAG members received this notification in their inboxes during the meeting (email Niccola Hobbs, DECC, 13:39 24/1/13). Whilst we understand DECC were under pressure to meet notification time scales, we do not believe due process was undertaken in this instance. ESTA decided not to object to the notification in the spirit of good relations and UK credibility, however it is important for DECC to accept that the proposal to use Zigbee at 2.4GHz was not fully endorsed by the stakeholders that they were consulting.

7. Access to IPRs

There are considerable Intellectual Property Rights (IPRs) that have built up over the two years Program development and whilst DECC assure us that contracted individuals pass all material that arises back to DECC, there will be IPRs that emerge that are less accessible, particularly when a larger group of individuals work for the same contracting organisation. There is a danger that there may be polarisation, especially if those organisations are already engaged with stakeholders in the Program. Attached to this evidence is the result of several freedom of information requests made to DECC, demonstrating the difficulty we have had in getting the information on contractors and the Company they work for. The head of end-to-end at DECC, who we originally thought was a civil servant is a contractor, and despite raising queries, we do not yet know who he works for, or who he reports to at DECC. A number of junior DECC staff were also under the misapprehension that the end-to-end design was being led by a civil servant.. This is a highly influential position and will define how the technology will emerge going forward, including participating parties and their costs. It must be impartial between supply and demand.

It is important that we know how many contractors plan to move over to the winning DCC bidder, and how much the procurement process depends on this. If it is too dependent, then it will be flawed, and we are in danger of spending public money on IPRs that we will never have access to.

8. Consumer Engagement

The Central Delivery Body (CDB), since it is paid for by all large suppliers is effectively being paid for by customers. This raises the question of why such expenditure is necessary. Granted, the trust element must be improved, but this is surely achieved by ensuring we have a good working meter market in which the consumer can operate. If, on initial installation, he doesn’t like metering offered by one supplier, he should be able to move to another, or a third party and be confident that the meter can be supported by a new supplier if it so wishes. This allows him to put pressure on the suppliers to innovate. It does not negate the universal SMETS specifications as the ideal; however ESTA believe derogation will be required as vendors start to deliver meters to their energy supplier customers. In addition, suppliers will wish to add value to their offering through unique metering features that others are not providing. The system must support this natural innovation process, allowing suppliers the option to take on new features inherited on CoS if they so wish. The current GBCS implementation approach will not allow them to do this—it forces a universal roll-out, which is proving impossible to achieve.

ESTA plan to be involved with the Smart Energy Code (SEC) and if possible to accede to the panel. DECC have provided cost figures for the daily collection of half-hourly data which are far less than those ESTA members can currently achieve. It means that we can provide Energy Automatic Monitoring and Targeting Systems (AM&T) to a much wider audience (lower consumption thresholds), and it may even be that this model drives adoption of half-hourly settlements in the domestic sector. This is something the economists have wanted for some time, but it has been resisted by suppliers because of the systems they would need to process it. We hope that DECC will continue to encourage this initiative from ESTA.

9. Stranded Assets

DECC have a duty to ensure that meters installed under the smart metering initiative do not need to be removed before the end of their life. Interoperability is essential to facilitate this. Further work must be done to identify how the 600,000 non-SMETS and SMETS1 meters can remain on the wall and continue to operate on CoS. Vendors of these meters must be invited to offer their protocols and other details to the industry so that solutions can be best explored. Criteria for roll-out obligations must be reviewed. If DECC envisage only a single register read will be achieved interoperably before SMETS2, then it must allow all meters that provide this, and meet European technical metering requirements (MID) to count against roll-out. Alternatively DECC must not allow further meters that are not fully interoperable to be installed before DCC is ready to receive them. In this case, the risk of advanced adoption must clearly remain with those taking the risk, and customers switching supplier should not have to pay for a meter a second time. The payment should come solely from the supplier that took the risk. The alternatives are clear. Either the advanced adopters make provision for their technology to be accessible outside their supply chain, or they pay for the consequences of their equipment not being compatible. The charges must be transparent to the consumer so that he knows the cost implications of meter replacement against what he was originally offered (n.b. It may have been a “free” offer of course).

Another alternative would be to insist that the DCC makes arrangements to adopt these assets (ie operates the necessary head-ends and comms). ESTA originally understood this to be mandatory, but believe this requirement has now been relapsed. This will avoid SMETS2 meters bulldozing out existing advanced meters in PC3&4 to the detriment of ESTA members existing working products and service.

10. Environment and Sustainability

It is noticeable that environmental and sustainability groups have been absent from the SMETS development process so far. If a good working meter market can be achieved, then this is not a critical factor because influence from these groups will be included through natural market forces. However in the meantime ESTA do recommend closer engagement with such parties, and appreciate the time DECC have provided in ESTA face-to-case meetings over the last year.

June 2013